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 settlegit 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/settle)<a href="https://agentmods.dev/skills/michaelycjo/specseal/settle"><img src="https://agentmods.dev/badge/skills/michaelycjo/specseal/settle/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/settle"><img src="https://agentmods.dev/badge/skills/michaelycjo/specseal/settle.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 | $0.00113 | $0.04412 |
| Opus 5.5 | $0.00045 | $0.01765 |
| Sonnet 5.5 | $0.00023 | $0.00882 |
| Haiku 4.5 | $0.00011 | $0.00441 |
Grade A, and why
settle 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 — 336 lines — stays where its author put it; the contents beside it link to each section on GitHub.
settle — the fold a shipped spec has been waiting for
A work item's directory is written to be read while its pull request is open.
After the merge nothing mechanical reads it and people rarely do, and it stays
on disk forever: 98 work items, 1,338 files, 15M on the tree this shipped
from. seal/README.md has said each one "waits until a later settle folds
it" since the root existed, and until now nothing folded anything.
The command reads and groups; you judge and write.
docs/one-root-by-lifetime.md §What keeps settle light fixes the split —
the step "moves and does not verify" — and folding only what is still true
is a judgment about truth. So settle supplies the corpus and never writes a
sentence into docs/; every standing statement below is written by the
session, and the command's second arm removes only what one of them already
absorbed.
settle what would fold, grouped by segment
settle --retire remove the directories whose fold docs/ records
settle --released-at REF what counts as released, and the base the rule is
asked at (default origin/main)
The script is skills/settle/scripts/settle.py, and bin/settle is on the
Bash tool's PATH while the plugin is enabled. It needs python 3.12 or newer
and says so at entry rather than dying partway through.
It refuses to run in local mode, and that is the one layout it will not
touch. A seal/ root under the common git directory is never committed, so
no ref holds the work item directories, nothing in them reads as released, and
nothing removed from them could be recovered. seal mode shared moves the
root into the tree and settle works from there. Every other command in this
plugin reads both places; this is the one that stops, because it is the one
that deletes.
When it runs
By hand, named as a step in the release checklist. It is invoked, never triggered: no hook runs it, no gate refuses a branch over an unfolded work item, and it is not part of the release-preparation commit that gathers the changelog and ledger fragments. Folding writes policy prose, which is a judgment act, and a release that stops for somebody to write documentation is a release that stops.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 Changed · +2 lines 7827e01cd486
- 4d ago Changed · +37 lines fd21aedc4ddc
- 5d ago Changed · +99 lines ad5489b4c73a
- 6d ago First seen · 198 lines · 113 tokens per session scan A 61af632435a1
settle is a skill published in the GitHub repository MichaelYcJo/SpecSeal (1 stars, last pushed today), licensed MIT. It adds 113 tokens to every session and 4,412 once invoked, about $0.0005 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 2026-09-23.
Other skills, from other repositories
work-story
Work a user story end to end — fetch the ticket, plan (with approval), implement, verify the applicable gates, self-review, open the PR, and update the tracker. Use when given an issue key like PROJ-1234, ENG-42, or.
procoder
Work like a senior developer in a repository governed by procoder: run the commit gate before calling anything done, format and lint through the binary, and drive the spec, plan, todo, backlog, and sprint chain in .procoder/. Use this skill when the repository contains a .procoder/ directory or an AGENTS.md naming…
implement
Use when the user wants an existing GitHub issue turned into a finished, reviewed pull request — "implement issue 123", "pick up ticket 45", "build the feature described in issue 7", "go fix the bug in issue 99", "start working on that issue". Triggers on any request to do the work an issue already describes, whether…
pr-feedback
Work a reviewer's comments on a pull request to the end - all three comment surfaces enumerated before any is triaged, every comment fixed, declined with a reason, or ticketed, push before you reply. Use when a PR comes back with review feedback or a red check.
loop-gh-pr
A recipe for a scheduled GitHub pull-request watcher. It checks open pull requests for new activity and prepares read-only review drafts without posting comments or merging anything.
loop-bb-pr
A command that prepares a repeating check for a Bitbucket pull request. A pull request is a request to review and merge code changes into a project.