Codexa

Codexa is a local change-evidence layer for human and agentic software workflows: committed-change receipts, plan conformance, blast-radius review, and verification crediting built on a deterministic codebase map.

In a pull request or terminal, Codexa reviews the committed base-to-head range and produces one receipt for developers, CI, and automation. If Codex or Claude Code is the editor, the same engine also shows what the agent read, what plan it saved, what changed, which checks would earn verification credit, and which gaps still need an honest handoff.

In plain English: it reads a repository, builds a compact index of the files, symbols, imports, tests, risks, workflows, process traces, and graph clusters it can prove, then gives Codex, Claude Code, or another MCP client small evidence-backed packets when direct source inspection is not enough. Exact, local work stays source-first instead of paying a mandatory context-tool tax. Codexa is meant to help an agent answer questions like:

• What should I read first?
• What could this change break?
• Which tests are relevant?
• Did my final dirty tree match the plan I saved before editing?
• Did the verification commands the agent reported actually prove anything?
• What does this committed branch change, and what should a reviewer verify?

It is not an autonomous coding agent. It does not edit your source files through MCP. It is a context compiler, query server, and verification guide.

Why Codexa

Seven capabilities are deliberately hard to find elsewhere:

• One committed-change receipt. codexa review resolves an explicit Git base and head without a shell, requires the head to match a clean indexed checkout, and returns bounded identity, file changes, diff statistics, graph impact, plan conformance, test recommendations, supplied execution reported verification claims, verdict, and next actions. The CLI, GitHub Action, and advanced MCP change_review operation use the same structured result. Observe mode is the default and never blocks on heuristics; explicit fail mode blocks only on local, range-bound plan drift or reported command failures.

• A drift loop. change_plan snapshots per-file hashes plus symbol and risk baselines before editing; post_edit_review diffs the real dirty tree against that plan afterwards, rename-aware. When no plan was saved, the pre-edit hook saves an implicit baseline automatically, so the review always has a pre-edit reference; an explicit change_plan upgrades it with planned scope, tests, and explicit task invariants. Distinct repeated attempts are counted by task and plan revision; a mandatory replan remains latched until the agent saves a newer accepted plan. Blocking is opt-in: only reviews against an explicit plan can surface a blocking verdict to the host — implicit baselines keep the loop informational.

• Exact checkout identity. Every query validates that the index belongs to the selected canonical worktree and current HEAD. A mismatched checkout, copied index, changing Git probe, or dirty overlay that changes while an authoritative result is being persisted fails closed. A stable dirty overlay remains valid input to change planning and post-edit review; auto-refresh gets one repair attempt and must pass the same identity check before any context is returned.

• A verification ledger. Commands the agent reports are parsed against a faithful POSIX-shell subset before earning coverage credit: npm test || true earns nothing, tsc --help is vetoed as non-compiling, sh -c wrappers are unwrapped with ambiguity failing closed. Scope stated plainly: this detects structural exit-masking in reported commands — it cannot detect a wholesale fabricated report. Targeted Playwright Test files receive reported credit across direct, launcher, and package-script forms; unscoped, project-only, list, UI, and zero-test-tolerant invocations stay uncredited. The opt-in AutoVerify lane exists for execution-backed evidence. Coverage, ledgers, and proof cards label that difference explicitly: executed-by-autoverify evidence ranks above imported live-run manifests and reported commands, while unauthenticated imported artifacts remain explicitly reported rather than being presented as witnessed execution.

• Compaction-safe continuity. Session decisions, rejected hypotheses, invariants, run-artifact references, and stopping conditions are carried by bounded task/session state. Compaction archives are published before active detail is removed, and proof cards verify snapshot memory pointers against the active store or the bounded archive.

• Graph-aware relational packets. v0.7.0 precomputes bounded process packets, functional module clusters, graph-view exports, and opt-in summary prompts. search now reports raw exact-hit counts beside Codexa-ranked anchors, so a "zero grep hits" task can still surface one ranked symbol or file target plus related process and cluster context.

• A fail-closed eval. The eval harness runs real rg/git baselines and fails a scenario outright if the raw baseline does the job better. The archived v0.2.0 release run passed 20/20 scenarios with packets averaging 0.66x the raw baseline output size — and the harness ships in this repo, so you can re-run it yourself. See Public Proof.

Limits, stated up front: TypeScript/JavaScript and Python are the deep lanes (Rust/Go/Java are shallow; other languages get light file facts). Impact expansion caps at graph depth 3. The tested envelope is repos around the ~50K-LOC scale of Codexa itself — expect slower cold indexing and shallower ranking on large monorepos. Everything runs locally: zero API keys and zero network calls in the core paths.

Maintainer Expectations

Codexa is maintained by one person, in spare time, with a deliberately narrow scope. That shapes how this repo works:

• Response times are days to weeks, not hours.
• Scope is narrow on purpose. Deep native language indexers, new LLM analysis layers, broad IDE products, and general-purpose search modes are usually out of scope.
• Not every working PR will be merged. Open an issue first for anything beyond a typo or small docs fix.
• Security issues go through private advisories, not public issues. See SECURITY.md.
• Questions and "is this the right tool?" discussions belong in Discussions, not the issue tracker.

Quick Start

Codexa requires Node.js 22 or newer.

Install from npm:

npm install -g @mirnoorata/codexa

Or work from a checkout:

git clone https://github.com/mirnoorata/codexa.git
cd codexa
npm install
npm run build
npm link

Wire Codexa into another repository:

codexa init /path/to/project --policy-pack            # Codex CLI: .codex/config.toml + hooks + local proof policies
codexa init /path/to/project --claude --policy-pack   # also writes a repo-root .mcp.json for Claude Code
codexa init /path/to/project --ci                     # also writes a read-only pull-request review workflow
codexa session-start /path/to/project
codexa prove /path/to/project --task "make this change safely"

For a start-to-finish first setup, see the new user tutorial. For a host-focused install choice, see No-Brainer Install Guide.

After codexa init, the target repository gets a repo-local .codex/config.toml entry that lets Codex discover the Codexa MCP server automatically, and with --claude a repo-root .mcp.json so Claude Code discovers the same server (only the codexa entry is managed; other servers in an existing .mcp.json are preserved, and malformed JSON aborts the write). When init runs from an evictable npx cache, generated configs pin npx -y @mirnoorata/codexa@<version> instead of the cache path so they keep working after a cache prune.

Linked git worktrees are wired the same way. Untracked .codex/config.toml and hook files stay host-local, so a fresh worktree is invisible to Codexa until you run init in it. If a team intentionally tracks those files, init renders worktree-relative launch commands and keeps the shared files unchanged when the same branch is checked out at a different path:

git worktree add ../my-feature feature-branch
codexa init ../my-feature        # non-interactive: config + hooks + a fresh index for the worktree

The worktree gets its own index (its HEAD and dirty state differ from the parent checkout's, so the parent's index would serve stale answers). If you automate worktree creation, add codexa init to that automation; tracked wiring remains Git-clean while the worktree-local ignored index is refreshed.

Codexa binds an index to the canonical worktree root, Git top-level root, HEAD commit, and workspace-state digest. Context and review queries fail closed when that identity does not match the active checkout. Auto-refresh may make one bounded repair attempt, but Codexa validates the rebuilt index again before serving an answer; --no-auto-refresh never serves a mismatched index.

Useful flags: the default tool profile for fresh installs is core — only search, change_plan, and capabilities are advertised directly, which reduces the decoded tools/list JSON surface. The compact capabilities dispatcher keeps every non-core operation reachable through the same operation-specific validation. --tools full also exposes every operation directly, and re-running plain codexa init preserves whichever profile the repo already uses. Fresh and core-profile Codex and Claude Code launches both pass serve --tools core, so the server enforces the compact surface even when a client ignores Codex's additional enabled_tools hint. --agents-md (opt-in) writes a managed Codexa workflow block into the repo's AGENTS.md for Codex, and --claude-md (opt-in) writes the same managed block into CLAUDE.md for Claude Code. The region between the <!-- >>> codexa managed --> / <!-- <<< codexa managed --> markers is reserved: Codexa replaces it in place on every re-run (so the block stays current) and never edits anything outside it. Unbalanced or malformed markers abort the write instead of silently truncating the file.

Committed change receipts

Use the receipt directly from a clean checkout. head must be the checked-out commit so Codexa cannot combine a different Git object with the current index:

codexa review . --base origin/main --head HEAD
codexa review . --base origin/main --head HEAD --format json
codexa review . --base origin/main --head HEAD --task-id my-saved-plan

The default --mode observe reports findings and exits successfully. warn uses warning annotations in GitHub output but remains non-blocking. fail returns exit code 2 only for range-bound local plan drift or a supplied structured command report with a nonzero exit; missing heuristic recommendations never become a blocking gate. Command and test claims are classified by the existing verification ledger and remain explicitly reported, not witnessed execution.

codexa init . --ci creates .github/workflows/codexa-review.yml. The managed workflow has only contents: read, disables persisted checkout credentials, checks out the exact pull-request head, and writes the receipt to the workflow summary and annotations. It does not comment on pull requests. Codexa refuses to overwrite a workflow it does not own. Re-running init --ci updates only the Codexa-managed workflow.

For portable plan comparison in CI, commit a redacted Codexa change-plan snapshot inside the repository and pass --plan-snapshot <path> (or the Action's plan-snapshot input). The loader accepts only a bounded, valid, non-symlink snapshot that resolves inside the repository. Local agent flows normally use --task-id or MCP change_review.taskId instead. A repository file is PR-controlled input, so portable plan conformance is advisory and never becomes a blocking verdict; local cache plans must also bind to the reviewed merge base before they can block explicit fail mode.

The installed command is codexa, and the server can also run ad hoc:

npx -y @mirnoorata/codexa serve /path/to/project --auto-refresh

Codexa is also listed in the official MCP registry as io.github.mirnoorata/codexa for MCP clients that discover servers there.

For shared workspace launches such as codexa serve /srv, Codexa can route to the active project recorded in .codex/WORKING.md. Selected session rows win; conflicting active focus, workspace default, or active-session evidence fails closed instead of silently choosing the wrong repo. Use CODEXA_WORKSPACE_SESSION=<session> or --workspace-session <session> when serving a shared workspace root with multiple live workers.

Proof cards and policy packs

codexa prove is the compact "should I trust this agent handoff?" view:

codexa prove /path/to/project --task "change auth timeout behavior" --diff

It reports:

• index freshness and current dirty-tree state;
• read-first files selected from the task and graph context;
• saved change-plan snapshot status, including planned edit targets, planned tests, and exact task invariants when a snapshot exists;
• current task-lifecycle state, including any latched mandatory-replan stop;
• a bounded decision log recovered from active session memory or its compaction archive, with pointer-integrity diagnostics;
• verification commands, ledger preview, and reported commands/tests/reports classified with the same command-credit rules as post-edit-review;
• explicitly selected, immutable live-run artifacts bound to the exact task, HEAD, and workspace-state digest;
• explicit trust tiers on coverage and ledger rows, so an agent-reported pass cannot look equivalent to a fresh AutoVerify execution;
• local policy-pack status and remaining proof gaps.

codexa policy-init /path/to/project writes a small local policy pack under .codex/policies/ (verification.json, complexity.json, security.json). codexa init /path/to/project --policy-pack creates the same pack during initial setup. The files are plain JSON, are not executable, and are consumed by codexa prove as bounded local evidence. Neither init nor policy-init overwrite existing policy files unless policy-init --force is passed.

Works with any MCP host

Codexa is deterministic and model-agnostic — its core indexing, ranking, and query paths call no model and need no API keys, so it serves the same evidence-backed context to any agent host that speaks MCP: the OpenAI Codex CLI (repo-local .codex/config.toml), Claude Code (codexa init --claude writes a repo-root .mcp.json; the bundled plugin under integrations/claude-code/ ships its own MCP server entry, hooks that auto-save the pre-edit baseline and surface blocking drift verdicts to the model, and slash commands; --claude-md adds workflow steering — pick the plugin or init --claude for MCP wiring, not both, or Claude Code will register the codexa server twice), and any client that discovers it through the MCP registry. There is no per-model integration to do — the model lives in the host, and Codexa is the host's context server. (The one exception is the opt-in, off-by-default semantic lane, which can call a configured embedding provider such as OpenAI — see Optional Lanes.)

The Codex plugin bundle under plugins/codexa/ ships an MCP wrapper and skill, not a post-edit hook. codexa init adds edit-scoped Codex hooks, but those run before later shell verification and do not claim final completion ownership. The Claude plugin's Stop hook remains a useful advisory drift gate, but it does not claim evidence-bearing completion ownership: the agent retains one final post_edit_review route for actual command reports and invariant reviews.

Result-size discipline is built in: every tool description states its typical output size, and structured results are budget-compacted with truncation records naming dropped fields. CODEXA_MCP_STRUCTURED_BUDGET_BYTES caps the serialized UTF-8 JSON bytes of the structuredContent.data subobject; mandatory envelope identity, lifecycle, and resource metadata is additional, so telemetry's totalBytes measures the complete reserialized result object. Analysis tools whose evidence can expand accept responseFormat: "auto", "concise", or "detailed"; the already-bounded freshness result has no format switch. Automatic and explicit concise delivery always return a bounded concise receipt. When available, an immutable resource URI preserves the bounded detailed packet. If omitted detail is required for a safe decision and that resource cannot be persisted or retrieved, the concise receipt becomes blocked and directs the caller to request explicit detailed output. Explicit detailed output is inline but still subject to a hard total-result bound. The tools/list surface is budgeted too: the per-tool output schema defaults to a compact top-level contract (CODEXA_MCP_OUTPUT_SCHEMA=full restores the deep schema). Managed and bare codexa serve launches default to core, so manual MCP entries receive the same bounded surface without a migration step. Use --tools full only when an older client requires direct names for every operation. Core advertises three direct tools while retaining the same logical operations through capabilities.

The deterministic transport benchmark reserializes decoded JSON and counts UTF-8 bytes. That is a provider-agnostic proxy, not a measurement of model input tokens, price, or provider-specific wire serialization. Its clean fixture supports a reduction in tool advertisement plus capability-discovery bytes and the hard worst-case result budget. Ordinary first and repeated task-result bytes were unchanged, so Codexa does not claim that core mode makes ordinary result packets smaller.

Managed cloud agents

Codexa's stdio transport is for a host running on the same machine as the repository (Codex CLI, Claude Code). Its HTTP transport is loopback-only by design — non-loopback bind addresses and non-loopback Origin headers are rejected — so a hosted agent whose container runs in someone else's cloud (for example a Claude Managed Agents session) cannot reach a local Codexa server over the public network.

The supported way to give a managed cloud agent Codexa context is a self-hosted sandbox: run the agent's tool-execution container in your own infrastructure, alongside a Codexa server, and point the agent's MCP config at Codexa on 127.0.0.1. The agent loop stays on the provider's orchestration layer; tool execution — and the Codexa connection — stay inside your trust boundary, where loopback HTTP is safe. An authenticated remote HTTP mode that would let a provider-hosted container dial into Codexa directly is intentionally not shipped: exposing a codebase context server to the network needs an auth/origin policy Codexa does not yet have, so it is deferred rather than shipped insecure.

The Everyday Workflow

Use Codexa selectively as a guardrail around code changes. A normal bounded agent task should usually use no more than two Codexa calls. Exact local work may use zero. The narrow three-call safety exception is an ambiguous, materially risky edit in a host with no completion/Stop gate: search -> change_plan -> post_edit_review.

1. Start with source tools when the target is exact and local. Read the named files or symbols and use repository-native verification with zero Codexa calls. A raw-sufficient search result is terminal for discovery: work from the exact hits instead of requesting another context packet. A materially risky edit may still warrant one change_plan after the source has established the target.

2. Spend one discovery call only when the target is ambiguous. Use search, then stop if its raw results are sufficient. Use session_context instead for genuinely broad or resumed work; do not stack session_context, search, and task_brief for the same discovery need.

3. Save a plan only for a non-trivial or materially risky edit. change_plan with saveSnapshot=true, or CLI change-plan --save-snapshot, records intended scope, targeted tests, verification commands, and task invariants. For an ambiguous materially risky edit, search -> change_plan normally uses the usual two calls.

4. Edit and run the planned verification. Use the targeted tests and commands already returned by change_plan. Call test_plan only when that guidance remains unresolved or a dedicated verification plan is explicitly requested.

5. Let a true completion/Stop gate review after verification when one is installed and able to carry the actual verification evidence. Codex's codexa init hooks are edit-scoped and run before later shell verification; the Claude plugin's Stop hook also lacks a trusted command/invariant ledger. Both retain one final evidence-bearing review route. Without a qualifying completion/Stop gate, an exact materially risky task may use change_plan -> post_edit_review. If the target was also ambiguous, the safety-preserving sequence is the narrow three-call exception search -> change_plan -> post_edit_review. post_edit_review / post-edit-review compares the actual dirty tree with the saved snapshot, reports drift, checks declared task invariants, and tells you whether to continue, run tests, inspect, or replan. Repeated distinct attempts are accounted against a task-scoped loop budget; once the budget trips, the stop remains latched until a new saved plan revision is accepted.

6. Produce formal proof only when the handoff needs it. proof_card / prove binds policy changes, formal audits, releases, or artifact handoffs to freshness, a saved plan snapshot, task invariants, lifecycle status, local policies, and reported verification evidence.

When a managed pre-edit hook exists and you skip the explicit plan, it saves an implicit baseline of the dirty tree on the first edit. The review still gets changed-since-baseline and head-drift accuracy, but only an explicit plan enables unplanned-scope drift detection.

In core mode, use capabilities to discover or invoke a non-core operation; full mode also exposes each operation directly. Both paths use the same operation-specific schema and handler.

Automatic and explicit concise results stay within the concise result budget. When persistence succeeds, they also return a content-addressed URI for the bounded detailed packet under the active repository. The URI does not encode the checkout path, remains resolvable only for that server session, and is pinned by a durable live-session lease until that server shuts down. Concurrent MCP server processes share a hard per-repository ceiling of 256 result records and 256 session leases; a live owner's pins are never evicted merely because they are old. If a new unique pin, session lease, opaque route, or persistence write would exceed its bound, Codexa keeps the response concise. It remains self-contained when the retained receipt is sufficient; when omitted detail is required for a safe decision and no URI is available, it returns a blocked receipt that requests explicit detailed output. Graceful shutdown releases that server's leases; an abandoned lease is reclaimed only after its stale window and owner-process identity check. Unpinned records remain LRU-prunable within the 256-record disk bound. Explicit responseFormat: "detailed" returns a bounded detailed packet inline. Optional CODEXA_MCP_TELEMETRY_PATH records bounded mechanical byte/time events; byte accounting is synchronous, while file writes use a bounded queue off the response path. Graceful shutdown adds a content-free session-complete record; analysis excludes that footer from event totals and treats a missing footer as partial evidence. A relative telemetry path is resolved once against the configured MCP launch root, so a later workspace-focus change cannot split one server sequence across files. Each server session must use a unique destination that is absent at startup; the runner is responsible for enforcing that freshness precondition. The writer creates the path exclusively and leaves an existing path untouched, but an analyzer cannot infer from valid file contents alone which run wrote it. Telemetry never changes tool authority or completion scoring.

Selective MCP call budget:

exact/local/source-sufficient -> source tools, zero Codexa calls
ambiguous/raw-sufficient -> search, then stop
exact materially risky + completion/Stop gate -> change_plan(saveSnapshot)
ambiguous materially risky + completion/Stop gate -> search -> change_plan(saveSnapshot)
exact materially risky + no completion gate -> change_plan(saveSnapshot) -> post_edit_review
ambiguous materially risky + no completion gate -> search -> change_plan(saveSnapshot) -> post_edit_review

These are ceilings, not an automatic chain. Every additional call must be justified by unresolved ambiguity, material edit risk, or a missing completion review gate; a returned tool name alone is not a reason to keep calling Codexa.

What Codexa Builds

Running codexa index /path/to/project writes generated files under the target repo's .codex/codebase/ directory:

.codex/codebase/README.md
.codex/codebase/codex-contract.md
.codex/codebase/repo-map.md
.codex/codebase/relational-packets.md
.codex/codebase/relational-packets.json
.codex/codebase/relational-graph.json
.codex/codebase/packet-summary-prompts.ndjson
.codex/codebase/risk-map.md
.codex/codebase/placeholder-map.md
.codex/codebase/test-map.md
.codex/codebase/conventions.md
.codex/codebase/workflows.md
.codex/codebase/freshness.json
.codex/codebase/index.json
.codex/codebase/facts.ndjson
.codex/codebase/modules/
.codex/codebase/playbooks/

For lay readers, these are the maps and checklists Codex reads. For engineers, the durable machine-readable index is index.json plus facts.ndjson; the Markdown files are compact human/agent-facing projections of the same facts. relational-packets.md is the read-first graph packet view for process traces and module clusters; the JSON companions are bounded machine-readable exports for tools and graph visualizers. packet-summary-prompts.ndjson contains explicit opt-in prompt records only — indexing does not call a model.

Generated cache and working state live under .codex/cache/. Codexa-owned cache writes are allowed; source-file mutation is not exposed through MCP tools.

Main Commands

Most context commands auto-refresh stale or missing Codexa artifacts before answering. Use --no-auto-refresh when you intentionally want to inspect only the stored index.

What It Understands

Codexa indexes git-visible files and skips common generated or dependency directories. The source reader is intentionally small and deterministic.

Native parser lanes:

• TypeScript, TSX, JavaScript, and JSX through Tree-sitter plus TypeScript compiler assist.
• Python through Tree-sitter plus lightweight semantic assist.

Shallow deterministic lanes:

• Rust declarations, imports, methods, calls, and tests.
• Go packages, imports, functions, methods, types, constants, variables, and tests with module-aware import resolution.
• Java packages, imports, classes, interfaces, enums, records, methods, and direct call-like usage.

Lightweight file lanes:

• JSON manifests.
• Markdown, MDX, RST, and text docs.
• Shell scripts.
• Systemd service files.

Facts carry explicit confidence:

• authoritative: syntax or git facts Codexa directly read.
• derived: deterministic links, static assists, report-backed relationships, and likely test relationships.
• heuristic: framework hints, string references, dynamic behavior guesses, or risk hints.
• fallback: low-confidence context used only when nothing better is available.

Codexa should never make heuristic-heavy output look stronger than it is.

Architecture For Engineers

Codexa is a TypeScript package with five main layers.

1. Indexing

Entry point: src/indexer.ts.

Pipeline:

1. Discover git-visible files and dirty state.
2. Parse source files and reuse the content-hash parse cache where possible.
3. Import external static-analysis and symbol-report facts.
4. Apply TypeScript/Python semantic assists.
5. Resolve imports, usage sites, aliases, test edges, and graph links.
6. Rank files/modules with centrality, usage, churn, tests, dirty risk, and bounded outcome signals.
7. Build typed graph edges, workflow traces, functional clusters, and relational packet exports.
8. Record freshness, parser errors, and dirty hashes.
9. Publish artifacts atomically.

The indexer uses a cross-process cache lock so parallel Codexa commands do not stampede artifact writes.

2. Fact Model

Core types live in src/types.ts.

Important fact types:

• RepoSnapshot
• File
• Symbol
• UsageSite
• ImportEdge
• TestEdge
• GraphEdge
• WorkflowTrace
• ModuleCluster
• RiskSignal
• ParserError
• SessionMemoryEntry

Important graph edge kinds:

• DEFINES
• IMPORTS
• CALLS
• REFERENCES
• TESTS
• ROUTE
• JOB
• RISK
• ROUTE_HANDLES
• ROUTE_CALLS_STORE
• STORE_DISPATCHES_ADAPTER
• ADAPTER_REFERENCED_BY_MANIFEST
• UI_CALLS_ENDPOINT
• TEST_COVERS_WORKFLOW
• IMPLEMENTS
• EXTENDS
• EXPORTS
• TYPE_EXPORTS

Relationship claims can include EdgeEvidenceV1, which carries edge kind, source, confidence, reason, path/symbol endpoints, optional range, and stale/degraded flags.

3. Query Layer

Public query exports live in src/queries.ts, intentionally kept as a thin barrel. Implementations live under src/query/.

Key query modules:

• search.ts: repo maps, raw/BM25/exact/symbol/semantic search, target discovery, raw-exact-vs-ranked anchor reporting, and relational process / cluster packet selection.
• context.ts: context_pack, task_brief, focus_brief, and session_context.
• impact.ts: file/symbol blast-radius expansion and verification recipes.
• graph-traversal.ts: callers, callees, and dependency paths.
• workflow.ts: route/job/manifest workflow traces.
• change-plan.ts: pre-edit plans and saved snapshots.
• post-edit.ts: dirty-tree review against saved snapshots.
• test-plan.ts and tests.ts: test recommendations and provenance.
• verification.ts: command coverage, command envelopes, and verification ledger entries.
• session-memory.ts: cache-only working memory queries.

Query sessions (src/query/session.ts) carry the repo root, loaded index, freshness, git state, command budget, warnings, provenance, changed files, and changed symbols. Worktree inspection is allowed to degrade; an empty changed-file set with degradation warnings means "unknown", not "clean".

4. MCP Server

Entry point: src/mcp.ts.

Codexa registers a query-only MCP server. Stdio is the default transport for local Codex use. Streamable HTTP is available only on loopback addresses unless future auth/origin policy is added.

The default core profile advertises three direct tools:

search
change_plan
capabilities

capabilities dispatches every non-core operation through its own schema and handler. Full mode advertises the complete direct-tool surface:

freshness
repo_map
find_context
search
placeholder_report
symbol_context
impact
diff_impact
change_review
test_plan
task_brief
context_pack
focus_brief
session_context
callers
callees
dependency_path
workflow_path
change_plan
post_edit_review
proof_card
capabilities
session_memory

MCP resources expose generated .codex/codebase/ artifacts. MCP prompts expose small workflow prompts for impact-before-edit, dirty-diff review, snapshot edit loops, and targeted test planning.

MCP tools may update Codexa-generated artifacts or cache state when auto-refresh, snapshots, or session memory are enabled. They do not expose a source-editing tool.

5. Adapters, Packaging, And Release Tools

Adapters:

• src/cli.ts: Commander-based CLI.
• src/init.ts: repo-local MCP config and hook setup.
• integrations/claude-code/: Claude Code plugin, hooks, and slash commands.
• plugins/codexa/: hookless Codex plugin bundle with manifest, skill, and MCP wrapper; repositories initialized with codexa init supply the managed Codex hooks separately.

Operational tools:

• src/doctor.ts: local readiness checks.
• src/github-sync.ts: git/GitHub sync diagnostics.
• src/github-release.ts: release notes, tags, and GitHub Release flow.
• scripts/*.mjs and scripts/*.sh: source hygiene, privacy, package smoke, public snapshot, benchmark, and publish gates.

Optional Lanes

Semantic Retrieval

Semantic retrieval is opt-in and cache-based.

Build the cache:

codexa semantic-index /path/to/project --provider openai
codexa semantic-index /path/to/project --provider local-command --command ./embed-jsonl

After the cache exists, query commands can use it automatically when the snapshot and provider settings match. --semantic forces diagnostics, and --no-semantic disables the lane for one call.

OpenAI uses OPENAI_API_KEY and defaults to text-embedding-3-small. local-command receives JSONL on stdin and returns embedding records. Codexa does not ship a vector database and does not call embedding providers unless the semantic cache/provider path is configured or explicitly forced.

LSP Assist

LSP assist is read-only and bounded. Enable it with --lsp or CODEXA_LSP=1 on supported query commands.

Codexa can query:

• typescript-language-server --stdio
• basedpyright-langserver --stdio
• pyright-langserver --stdio

LSP failures are warnings in the packet, not hard failures. LSP never edits source files.

Static Analysis Reports

Codexa does not vendor Semgrep, CodeQL, ShellCheck, or other scanner engines. The default safe shape is report ingestion:

codexa static-analysis /path/to/project \
  --semgrep-report /tmp/semgrep.json \
  --codeql-report /tmp/codeql.sarif \
  --symbol-report /tmp/codexa-symbols.json \
  --scip-report /tmp/index.scip.json

Codexa also accepts a bounded CodexaSymbolReportV1 JSON document so external language tools can feed symbols and relationships into Codexa with derived confidence. SCIP reports are accepted as JSON exported by scip print --json; Codexa converts them into the same bounded symbol-report lane and does not run or vendor SCIP indexers.

Scanner execution flags such as --run-semgrep, --run-codeql, and --run-shellcheck are explicit opt-ins. They run installed local tools under scrubbed environments and write reports under `.codex/static-anal