history
The history of one item by its ID — its wording today, its records, the archive rows that mention it and every citation, gathered by a script; Claude adds a few sentences of its own
Install
npx skills add https://github.com/pche-broken-artist/forge-of-thought/tree/main/.claude/skills/history
claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install pche-broken-artist-forge-of-thought@llmmart
git clone https://github.com/pche-broken-artist/forge-of-thought.git
The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole pche-broken-artist/forge-of-thought collection as a plugin from our marketplace. Git is the plain clone.
Skill manifest
The history of one item (POS.1510, SOL.0630): what the project holds
of the item named by $0, printed in full for that one item and
nothing else. Read-only; this command writes nothing and runs
nothing but its script. This is the one way Claude reads an item's
history before proposing a change to it (CLAUDE.md, Versioning &
status): the log is searched by the script, never loaded whole into
the conversation.
- The project is the slug given as the second argument, else the
project in context (
forgeis the engine's own project); if that is ambiguous, ask. - Run
python scripts/forge-history.py <ID> <slug>(a short form such aspos580is accepted) and print what it printed, unshortened. What the script gathers, from where and in what shape is its header's;--fullprints every Was whole. - Add at most three sentences of Claude's own, marked as such: what the item says today, how it moved (born when, changed how many times, last at which version), and what hangs on it (a thread, a finding, a challenge among the citations). Nothing is derived from them and nothing is written: a change the reading suggests is proposed in the round that owns the item.
An ID nothing in the project knows is said so, as the script says it; the command never guesses a near ID.
Files (forge-of-thought)
-
SKILL.md 1.5 KB
--- description: The history of one item by its ID — its wording today, its records, the archive rows that mention it and every citation, gathered by a script; Claude adds a few sentences of its own argument-hint: "<ID> [project-slug]" --- The history of one item (POS.1510, SOL.0630): what the project holds of the item named by `$0`, printed in full for that one item and nothing else. Read-only; this command writes nothing and runs nothing but its script. This is the one way Claude reads an item's history before proposing a change to it (CLAUDE.md, Versioning & status): the log is searched by the script, never loaded whole into the conversation. 1. The project is the slug given as the second argument, else the project in context (`forge` is the engine's own project); if that is ambiguous, ask. 2. Run `python scripts/forge-history.py <ID> <slug>` (a short form such as `pos580` is accepted) and print what it printed, unshortened. What the script gathers, from where and in what shape is its header's; `--full` prints every Was whole. 3. Add at most three sentences of Claude's own, marked as such: what the item says today, how it moved (born when, changed how many times, last at which version), and what hangs on it (a thread, a finding, a challenge among the citations). Nothing is derived from them and nothing is written: a change the reading suggests is proposed in the round that owns the item. An ID nothing in the project knows is said so, as the script says it; the command never guesses a near ID.
Comments (0)
Sign in to join the conversation.
Reviews (0)
No reviews yet.
No comments yet.