Agent skill

evidence-lock

Write or rewrite a document in evidence-locked mode: no unsourced sentences — every substantive claim carries a footnote citing the exact passage in the user's provided sources, and anything unsupportable is explicitly marked. Use when asked to make a document fully sourced, add citations from my docs, ground a draft in the attached material, or produce something for audiences that will check (legal, board, regulators, enterprise buyers). Produces the document with numbered citations, a source map quoting each cited passage, and an unsupported-claims register.

mohitagw158561,255★ · 1 repos on radarProfile →
claude-codecursorMIT
Install
npx skills add mohitagw15856/pm-claude-skills --skill evidence-lock --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 1
SKILL.md size: 5 KB
Bundled scripts: none
Path: skills/evidence-lock/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 1,255
Language: HTML

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

From the SKILL.md

# Evidence Lock Skill For most drafts, plausible is enough. This mode is for the documents where someone *will* check: every substantive sentence either cites the exact passage in the user's sources that supports it, or wears an explicit `[UNSOURCED]` flag. No third state. ## What This Skill Produces - The **document**, with numbered footnote markers on every substantive claim - A **source map**: each footnote → source name + the supporting passage *quoted verbatim* - An **unsupported-claims register**: every `[UNSOURCED]` item with what evidence would resolve it - A **coverage score**: % of substantive claims that are locked ## Required Inputs Ask for (if not already provided): - **The sources** — pasted documents, files, or excerpts. This skill cannot run without them; general knowledge is not a source here. - **The task** — either a draft to lock (rewrite mode) or a brief to write from scratch (compose mode) - **Strictness** — *hard lock* (unsupported claims are removed to the register) or *soft lock* (they stay in the text, flagged `[UNSOURCED]`). Default: soft. ## Locking Method 1. **Index the sources first.** Skim all provided material and note what each source can support. I

What's inside
Steps it walks through
  1. What This Skill Produces
  2. Required Inputs
  3. Locking Method
  4. Output Format
  5. Quality Checks
  6. Anti-Patterns
More from pm-claude-skills
All skills →
About this skill
What does the evidence-lock skill do?

Write or rewrite a document in evidence-locked mode: no unsourced sentences — every substantive claim carries a footnote citing the exact passage in the user's provided sources, and anything unsupportable is explicitly marked. Use when asked to make a document fully sourced, add citations from my docs, ground a draft in the attached material, or produce something for audiences that will check (legal, board, regulators, enterprise buyers). Produces the document with numbered citations, a source map quoting each cited passage, and an unsupported-claims register.

How do I install it?

Run `npx skills add mohitagw15856/pm-claude-skills --skill evidence-lock --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From mohitagw15856/pm-claude-skills, a repository with 1,255 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going