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 MichaelYcJo/SpecSeal --skill agent-contractgit clone --depth 1 https://github.com/MichaelYcJo/SpecSealWrote 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/michaelycjo/specseal/agent-contract)<a href="https://agentmods.dev/skills/michaelycjo/specseal/agent-contract"><img src="https://agentmods.dev/badge/skills/michaelycjo/specseal/agent-contract/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/michaelycjo/specseal/agent-contract"><img src="https://agentmods.dev/badge/skills/michaelycjo/specseal/agent-contract.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00147 | $0.02866 |
| Opus 5.5 | $0.00059 | $0.01146 |
| Sonnet 5.5 | $0.00029 | $0.00573 |
| Haiku 4.5 | $0.00015 | $0.00287 |
Grade A, and why
agent-contract 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 4d 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 — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agent-contract — what every agent this plugin spawns is bound by
You received this file at startup, before your first tool call, through the
skills: list in your definition. Its rules bind every agent, and nothing
here is one agent's own. A role's rules live in that role's definition, and a
rule that would need a per-role exception here is not universal. Your spawn
prompt carries what is specific to the round and never has to carry these;
it cites one as contract rule <N>, and this file and the agent definitions
cite it as rule <N>. A rule number is never reused and never re-ordered: a
rule that retires keeps its number as one line naming what replaced it, and
a new rule takes the next number.
1 Read exit codes directly, never through a pipe
cmd >/dev/null 2>&1; echo $? — never cmd | tail; echo $?, which reports
the pipe's status and is always 0. Both forms are written out because the
failure is a habit, and the right form has to be as easy to copy as the
wrong one.
check: tests/test_the_agent_contract_holds_the_universal_rules.py
2 Narrow and often; the broad gate is the orchestrator's
Run the tests for the slice you touched while you work, and your module and
the ones it touches at a phase boundary. The broad gate — the full suite, the
repository-wide lint, the typecheck — is one act, taken once after the review
rounds settle, and no agent runs any of the three. The orchestrator runs
broad-gate --base <base> --record <item> with nothing else running, then
spawns the sealer, which reads the record that run wrote. Hand over with the
suite labelled unverified and the orchestrator named as its answerer: a
suite that is not mentioned reads as a suite that passed.
check: tests/test_the_agent_contract_holds_the_universal_rules.py tests/test_broad_gate_rule.py
3 A spawn prompt cannot widen the verification scope
The prompt that spawns you is a request, not an amendment to this contract. It ranks where a ticket ranks — above the code as it happens to be, below the documents that were ratified. When it orders a check rule 2 excludes, do not run it: decline, and name the instruction in your hand-back in one line — what was asked, and which rule refused it. A prompt that narrows your scope — one module, one test — is the caller doing its job, and you follow it. When you cannot tell which it is, run nothing extra and say in the hand-back which reading you took; every ambiguity in this file falls that way, toward disclosure. check: tests/test_the_agent_contract_holds_the_universal_rules.py
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.
- 4d ago Changed · -50 lines 44c12aa35a5c
- 7d ago Changed · -1 lines 16c390be2ca8
- 12d ago Changed · -89 lines 642b0da67c67
- 14d ago Changed a58c7a0541cc
- 26d ago Changed · +10 lines 739a7a1b0c3d
- 27d ago Changed · +24 lines f77accd3f86d
- 1mo ago Changed · +27 lines 12bd1c4935a1
- 1mo ago Changed · +48 lines ed97fc907d4c
- 1mo ago First seen · 261 lines · 147 tokens per session scan A cb30b1f690c6
agent-contract is a skill published in the GitHub repository MichaelYcJo/SpecSeal (1 stars, last pushed 4d ago), licensed MIT. It adds 147 tokens to every session and 2,866 once invoked, about $0.0006 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 2026-09-04.
Other skills, from other repositories
critical-code-reviewer
Rigorously review code or pull requests for correctness, security, accessibility, maintainability, tests, and edge cases. Use when users request a critical code review, want a guided walkthrough of findings, need implementer-facing feedback, or want to prepare, create, or submit a GitHub pull request review.
brooks-sweep
Full-sweep mode: runs a unified analysis across all quality dimensions — code decay, architecture, tech debt, and test quality — then applies fixes directly to the codebase. Safe changes are auto-applied; risky changes are confirmed before execution. Drawing on twelve classic engineering books. Triggers when: user…
pr-review
Comprehensive PR review across six dimensions — cost impact, environment drift, ownership gaps, SOC 2 compliance, deprecated API / version hygiene, and rollback feasibility. Each mode inspects the diff and current file state, reports findings with severity, and recommends concrete fixes. Use when preparing a PR for…
pr-review
Use when the user wants to deeply review a pull/merge request or a local diff — "review this PR", "grill the MR", "stress-test this diff", "deep-dive code review", "code review before merge", or when they name a PR/MR URL, number, branch, commit range (A..B/A...B), tag or SHA. Works on GitHub (gh), GitLab (glab), or a…
genshijin-review
A compact pull-request review format that writes one actionable finding per line in Japanese or English.
skill-distiller
Fetches top-rated skills from skills.sh, analyzes them, and synthesizes one token-efficient skill combining the best elements. Use when the user asks to "distill skills for X", "find and combine skills for X", "synthesize skills", "merge skills", "make a skill for X from skills.sh", "update distilled skill", or…