agentleFS
Sign inStart free
Cor-IncorporatedCor-Incorporated/claude-code-skills/skills/context7-skills/SKILL.md

context7-skills

Skill5 starsChanged 2 months agoMITView on GitHub

Manage Context7 CLI skills: search, install, list, remove, info via ctx7. Use when the user wants to find, install, list, remove, or inspect Context7 skills. Triggers: 'search skills', 'install skill', 'list my skills', 'remove skill', 'ctx7', 'context7'. Do NOT use for general package management (npm, pip), non-Context7 CLI commands, or skill authoring.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 8 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No other commands may be executed.

    line 21
  2. Distinctive:

    If multiple targets are requested, stop and ask the user to pick one.

    line 33
  3. Very specific:

    Always execute via ctx7. The skills namespace is mandatory.

    line 37
  4. Distinctive:

    Only commands listed above may be executed.

    line 38
  5. Distinctive:

    If granted, execute immediately (no redundant user confirmation).

    line 47
  6. Distinctive:

    If denied, stop and tell the user to run locally or provide output.

    line 48
  7. Very specific:

    If execution fails with a network error (fetch failed, DNS, timeout), treat as environment limitation.

    line 49
  8. Distinctive:

    If the first run contains visible entries, do not rerun.

    line 54
  9. Distinctive:

    If invalid selection, ask for a valid number.

    line 61
  10. Distinctive:

    If no target was specified, prompt with numbered target list:

    line 62
  11. Very specific:

    Do NOT use --all for single skill install.

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Context7 Skills

Execute Context7 CLI commands directly. Printing commands without execution is forbidden.

Permitted Commands (Exhaustive)

IntentCommand
Searchctx7 skills search <keywords...>
Installctx7 skills install <repository> [skill] [--all] [target]
Listctx7 skills list [target]
Removectx7 skills remove <name> [target]
Infoctx7 skills info <repository>

No other commands may be executed.

Target Flags (exactly one allowed)

  • --global -- global
  • --claude -- .claude/skills/
  • --cursor -- .cursor/skills/
  • --codex -- .codex/skills/
  • --opencode -- .opencode/skills/
  • --amp -- .agents/skills/
  • --antigravity -- .agent/skills/

If multiple targets are requested, stop and ask the user to pick one.

Execution Rules

  1. Always execute via ctx7. The skills namespace is mandatory.
  2. Only commands listed above may be executed.
  3. At most one target flag per command.
  4. --all is only valid with install.
  5. Fix invalid, ambiguous, or incomplete input before execution.

Network Permission Gate

Commands requiring network access (search, info, remote install):

  1. Request outbound network permission from the execution environment.
  2. If granted, execute immediately (no redundant user confirmation).
  3. If denied, stop and tell the user to run locally or provide output.
  4. If execution fails with a network error (fetch failed, DNS, timeout), treat as environment limitation. Do not retry without explicit permission.

Search Output Rules

  1. Display results as a numbered list starting at 1. Preserve entry text as-is.
  2. If the first run contains visible entries, do not rerun.
  3. Rerun once only if: no visible entries appeared, or output is truncated with only summary lines.

Install Flow (from search)

  1. User replies with a number k to select a skill.
  2. Extract skill_name and repository from the selected entry.
  3. If invalid selection, ask for a valid number.
  4. If no target was specified, prompt with numbered target list:
    1. --claude 2. --cursor 3. --codex 4. --opencode 5. --amp 6. --antigravity 7. --global
  5. Do NOT use --all for single skill install.
  6. Request network permission if remote, then execute: ctx7 skills install <repository> <skill_name> <target_flag>
  7. Display raw CLI output as-is.

Error Handling

  • Missing skills namespace: add it automatically and proceed.
  • Command outside permitted scope: reject and list permitted commands.
  • More than one target flag: stop and ask user to pick one.
  • Network failure: report as environment limitation, do not retry.
  • Empty search results: suggest broadening keywords.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free