Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add markmhendrickson/neotoma --skill reviewgit clone --depth 1 https://github.com/markmhendrickson/neotomaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/markmhendrickson/neotoma/review)<a href="https://agentmods.dev/skills/markmhendrickson/neotoma/review"><img src="https://agentmods.dev/badge/skills/markmhendrickson/neotoma/review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/markmhendrickson/neotoma/review"><img src="https://agentmods.dev/badge/skills/markmhendrickson/neotoma/review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00057 | $0.06257 |
| Opus 5.5 | $0.00023 | $0.02503 |
| Sonnet 5.5 | $0.00011 | $0.01251 |
| Haiku 4.5 | $0.00006 | $0.00626 |
Grade A, and why
review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 366 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review
Perform a structured code review of a GitHub PR, a branch, or the current working tree against a base ref.
Invocation forms
| Form | Meaning |
|---|---|
/review |
Review uncommitted + committed changes on current branch vs main |
/review <base>..<head> |
Review a specific ref range (e.g. v0.13.0..HEAD, main..feature/x) |
/review PR#N |
Fetch PR diff via gh pr diff N and review it |
/review --base <ref> |
Override the base ref; head defaults to HEAD |
When to invoke
- Before merging any feature branch to
main - As part of
/releaseStep 3.6 (test coverage review) on the full release diff - When changes went directly to
mainand need a retroactive review pass - When asked to review a specific PR or diff range
- From
run_feature_workflowbefore the final merge step
Workflow
Phase 1 — Scope the diff
-
Determine base and head refs from the invocation form above.
-
Run:
git diff --stat <base>..<head> git diff --name-only <base>..<head>For a PR:
gh pr diff <N> --name-onlyandgh pr view <N>for metadata. -
Compute surface summary:
- Files changed, lines added/removed
- Which surface categories are touched (data layer, API/contract, auth/security, CLI, tests, docs, release, observability)
- Whether any high-risk surfaces are present (destructive ops, file-shape parsers, new CLI commands, HTTP runtime config, auth middleware)
-
Print a one-line scope summary before proceeding:
"Reviewing .. — N files, +X/−Y lines. Surfaces: [list]. High-risk: [yes/no]."
Phase 2 — Load docs
Always load (3 docs — true invariant kernel only):
docs/NEOTOMA_MANIFEST.mddocs/foundation/layered_architecture.md.claude/rules/change_guardrails_rules.md
Load conditionally based on paths in the diff (per docs/developer/pr_review_reading_list.md):
| If diff touches | Also load |
|---|---|
src/reducers/ |
docs/subsystems/reducer.md |
src/services/ingestion*, src/services/store* |
docs/subsystems/ingestion/ingestion.md, docs/architecture/idempotence_pattern.md |
src/services/observation*, src/services/sources* |
docs/subsystems/sources.md, docs/subsystems/observation_architecture.md |
src/services/interpretation* |
docs/subsystems/interpretations.md |
src/services/relationships* |
docs/subsystems/relationships.md |
src/services/entity*, src/services/schema_registry* |
docs/foundation/entity_resolution.md, docs/subsystems/schema_registry.md, docs/subsystems/schema.md |
src/services/entity_merge* |
docs/subsystems/entity_merge.md |
src/services/timeline*, src/services/events* |
docs/foundation/timeline_events.md, docs/subsystems/events.md |
src/services/search*, src/shared/action_handlers/entity_handlers* |
docs/subsystems/search/search.md |
src/services/deletion* |
docs/subsystems/deletion.md |
src/services/peer* |
docs/subsystems/peer_sync.md |
openapi.yaml, src/shared/contract*, src/shared/openapi* |
docs/architecture/openapi_contract_flow.md |
src/tool_definitions.ts, src/server.ts, docs/developer/mcp/ |
docs/developer/mcp/instructions.md, docs/developer/agent_instructions_sync_rules.mdc |
src/cli/ |
docs/developer/cli_reference.md |
src/actions.ts, src/services/local_auth*, src/middleware/ |
docs/subsystems/auth.md, docs/security/advisories/2026-05-11-inspector-auth-bypass.md |
src/actions.ts, src/server.ts, any auth/middleware path |
docs/security/threat_model.md |
| Guest access, public routes | docs/subsystems/guest_access_policy.md |
src/services/root_landing/ |
docs/security/threat_model.md |
New call to registry.register() (i.e. a new entity type being registered, not a schema field addition) |
docs/subsystems/record_types.md, docs/foundation/data_models.md |
Any if (entityType, switch (entity_type, or hardcoded entity-type list in diff |
docs/foundation/schema_agnostic_design_rules.md |
| Any new error code, tightened validation, or changed error response shape | docs/subsystems/errors.md |
Math.random, Date.now in non-observability paths, new sort/grouping in stored-output path |
docs/architecture/determinism.md |
| Any change to stored fields, observation shape, or entity snapshot output | docs/foundation/core_identity.md, docs/foundation/philosophy.md |
| Logging, metrics, tracing | docs/observability/logging.md, docs/observability/metrics_standard.md, docs/subsystems/privacy.md |
tests/ |
docs/testing/testing_standard.md |
docs/releases/ |
docs/developer/github_release_process.md |
package.json scripts |
docs/developer/package_scripts.md |
.claude/settings.json, *.sh, .github/workflows/ |
(no doc to load) — apply Phase 5 portability check |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 366 lines · 57 tokens per session scan A 9c93c3567903
review is a skill published in the GitHub repository markmhendrickson/neotoma (34 stars, last pushed 2d ago), licensed MIT. It adds 57 tokens to every session and 6,257 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
reviewer
Code review - correctness, tests, merge-readiness.
review-thread
Record review verdicts and run the StateNote review thread on an agent-operation instance — every ReviewRequest verdict is one atomic batch-direct-write that advances status and co-writes a reviewnote explaining why, and findings are corrected by appending notes rather than editing them.
aura-integration
Automatically captures Gemini CLI intent and writes it to .gemini.intent whenever code changes are made. Use this skill when making any modifications to the codebase so that the local Aura Semantic Version Control system can track the AI's reasoning.
plumb-line-audit
Use when auditing a diff or repository against the plumb-line principles — finds laundered uncertainty, boundary leaks, hardcoded priors, overstated maturity, outputs lacking recorded lineage, and baseline drift with no explanation. Read-only: it reports, never auto-fixes.
warrant-review
Review a pull request of s0fractal/warrant that the review loop says is due (its head closure is at the pull request's head, CI is green there, and no verdict stands), then file findings against that head closure or accept it, through the warrant MCP server, and push the records onto the pull request. Use when the…
cratis-engineering-review-rounds
Plan proportionate review rounds, distinguish blocking from advisory findings, and carry prior evidence into a delta round with a firm stopping point. Use when choosing a review round, handling blocking or advisory feedback, requesting a delta since the last reviewed SHA, or stopping recurring blockers. Not for review…