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 agentmods add skills/doidor/agentrig/self-verifynpx skills add doidor/agentrig --skill self-verifygit clone --depth 1 https://github.com/doidor/agentrigWhat 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 | $0.00046 | $0.00606 |
| Opus 5 | $0.00023 | $0.00303 |
| Sonnet 5 | $0.00009 | $0.00121 |
| Haiku 4.5 | $0.00005 | $0.00061 |
Grade A, and why
self-verify 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 2d 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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
self-verify (principle 5)
After producing changes, verify your own work before handoff. Do not invoke the reviewer until this loop converges.
Steps (do them in order; do not skip)
-
Baseline. Run the install/build/test/lint commands from
AGENTS.md'scommandsblock once before you make any edit related to the failing symptom. Capture the result:- For a fix scenario: confirm the suite is RED in the expected way (the target test fails).
- For a feature scenario: confirm the suite is GREEN (so you know your changes are what break it if it goes red later).
- Surface this baseline in your transcript — e.g. "baseline:
npm test→ 1 fail (divide-by-zero)".
-
Iterate. Make the change; re-run the commands. Cap at N=3 iterations.
-
After. Re-run the full suite at the end and surface the new state explicitly — e.g. "after fix:
npm test→ 0 fails, all 4 tests pass". The transition from baseline → after is the evidence that your work did what you claim. Reporting only "tests pass" without the baseline is half a self-verification. -
Self-park if still red. Leave a precise note (what failed, what you tried) and move the task to
parked. Never hand a red build to a reviewer.
Handoff checklist (run BEFORE you declare done)
- Baseline output captured + surfaced in transcript
- After output captured + surfaced in transcript
- Diff is on-target (no unrelated churn — check
git diff --stat) - Did you hit any non-obvious behavior or surprise? → run the
log-gotchaskill before handing off. This includes silently-passing-yet-wrong APIs, JS-floating-point quirks, framework defaults that bit you, environment surprises, etc. Wiki entries are how the next agent avoids repeating your discovery.
Notes
- Pin verification to your own HEAD; do not trust stale CI from an earlier commit.
- If the build is too expensive to run a full baseline (10+ min), at minimum run the smallest set of tests that demonstrates the symptom before AND after your fix.
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.
- 2d ago First seen · 50 lines · 46 tokens per session scan A bd7202272a8e
self-verify is a skill published in the GitHub repository doidor/agentrig (5 stars, last pushed 1mo ago), licensed MIT. It adds 46 tokens to every session and 606 once invoked, about $0.0002 per session on Opus 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.