recheck
Skill68 starsChanged 7 days agoMITView on GitHub
Re-verifies a finished Briefproof run after its files changed. A blind checker tests the requirements whose evidence files moved, the ledger records what still holds and what broke, and the proof snapshot is retaken so the pane's warning clears. Starts only on /briefproof:recheck, usually from the pane's Re-check button.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- zahidakhyar/skillstheir own version, first seen Nov 1, 2026
- winchxyz/idea-to-buildtheir own version, first seen Jun 9, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- winchxyz/idea-to-build★ 9same name
- zahidakhyar/skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 54
Do not ask whether to drop it: the user asked for it once, and it held.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Briefproof · re-check
First, check you are in Claude Code. This skill needs the plugin's ledger tool,
mcp__briefproof__state. If it is not available, tell the user in one line,
"Briefproof runs in Claude Code: open the Code tab (local session) or a terminal
and run /briefproof:recheck", and stop.
A re-check answers one question: does the project still do what the last run
proved? It builds nothing and changes no source file. Fixing is
/briefproof:change.
Start
state {action: "show"}. It prints the run folder, {run}, and, when files
moved, a drift: line naming the changed files and the requirements that depend
on them (R03, R05 may be stale; R02 (base) means a requirement of the run
this one was a change to). If there is no finished run, say so and stop. If
show prints no drift: line, nothing changed since the last check: say so in
one line and stop, unless the user asked in words for a full re-check; then every
live requirement is on the list.
state {action: "recheck"}. From here the ledger routes your requests and the
checker's to the Acceptance model and counts them there; the pane shows the
re-check in progress. (The pane's button has usually done this already; calling
it again is harmless.)
Check
Clone the project to a temp dir (git clone, or copy without node_modules,
.venv, .git). Spawn a fresh subagent with prompts/recheck.md, giving it
only {run}/1-brief/brief.md, the clone, and the quotes of the stale
requirements. For a change run, the brief already carries the ## Change
request section; for a base requirement, take its quote from
{run}/1-brief/base-requirements.md.
Save its reply as {run}/7-acceptance/recheck-<YYYYMMDD-HHMM>.md; never
overwrite acceptance.md.
Record
Per requirement the checker reported on:
✓→set_requirement {id, status: "done", files: [<its FILES>], note: "<evidence>"}. Pass the files even when they did not change; the new list replaces the old one. Paths are relative to the project root; the ledger refuses absolute ones.✗or~partial→set_requirement {id, status: "broken", note: "<evidence>"}. Do not ask whether to drop it: the user asked for it once, and it held; now it does not. The pane shows it as "no longer holds" until a change run fixes it.- A base requirement (
R02 (base)) has no id in this run. If it still holds, nothing to record. If it broke:add_requirements {quotes: ["<its quote>"]}, thenset_requirementon the new id withstatus: "broken", the files and the notebase R02 broke: <evidence>. It is now this run's problem to show.
Then state {action: "snapshot"}. The ledger hashes the evidence files as they
are now, this run's and its base's; the pane's warning clears and the report
page is rewritten.
Report
Two to five lines in the chat: how many were checked, which still hold, which
broke (with the evidence in a few words), and, if anything broke,
/briefproof:change <what to fix> as the next step. No file needs writing beyond
the recheck reply.
Red flags
- You fixed something. A re-check only looks.
- You marked a requirement
donethat the checker did not see work. - You skipped
snapshot: the warning stays, the Acceptance model stays in force, and the next session repeats the work.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

