agentleFS
serdardbserdardb/context-bridge/codex/SKILL.md

bridge

Skill4 starsChanged 58 days agoMITView on GitHub

Hand this session off to another coding agent via context-bridge (usage: $bridge claude, $bridge grok, $bridge antigravity, $bridge opencode)

5 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions5 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 5

    1. msanchezdev/agent-bridge★ 20same name
    2. EthanSK/agent-bridge★ 11same name
    3. pantheon-org/tekhne★ 10same name
    4. ChainGPT-org/chaingpt-claude-skill★ 7same name
    5. tyler-berggren/tb-claude-kit★ 2same 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

7 rules. 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Follow these steps exactly:

    line 8
  2. Distinctive:

    Read the requested target agent from the invocation.

    line 10
  3. Distinctive:

    Write a SUMMARY of this session for the agent taking over, covering only what happened since the last bridge point.

    line 13
  4. Distinctive:

    Run exactly this shell command, with the requested agent as the target:

    line 38
  5. Distinctive:

    If the command fails because the summary is too large, it says so and gives both numbers.

    line 45
  6. Very specific:

    If the command fails AND its error output mentions --adopt, an unlinked session was discovered heuristically: relay the error's timestamps …

    line 51
  7. Distinctive:

    If it succeeds, reply with exactly one short sentence like "Handing off to Claude — the bridge will switch automatically." and END YOUR …

    line 58

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “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

The user invoked the context-bridge handoff, e.g. $bridge claude, $bridge grok, $bridge antigravity or $bridge opencode.

Follow these steps exactly:

  1. Read the requested target agent from the invocation. If none was given, ask which agent to hand off to and stop until they answer.
  2. Write a SUMMARY of this session for the agent taking over, covering only what happened since the last bridge point.

    The bridge already carries the transcript, the git changes and a record of every command that ran. What it cannot carry is which of that mattered, and you are the only thing that knows. So do not list what happened; say what it means.

    Cover what was decided and why, what was tried and rejected, what is in flight, and what is unresolved. Say what the next agent should verify first. Where you are not certain, say so rather than writing it as settled: a summary that sounds equally sure of everything is worse than one that marks its own soft spots, because the reader cannot tell which parts to check.

    Length is yours to judge. There is no item count and no sentence limit. The only limit is a byte budget, and the command will tell you the number if you go past it. Do not summarise the whole project, only what happened since the bridge brought you here; if nothing of substance happened, say that in a line.

  3. Compose two short notes as well:
    • DECISIONS: concrete decisions made, including rejected approaches and why.
    • NEXT: the current objective and unresolved questions. Both semicolon-separated. These are the scannable form of what the summary says at length, not a replacement for it.
  4. Run exactly this shell command, with the requested agent as the target:

    bridge handoff <target> --summary "<SUMMARY>" --decisions "<DECISIONS>" --next "<NEXT>"

    The bridge itself decides which agent is handing off, which target is valid, and how the context travels. Do not add flags of your own.

  5. If the command fails because the summary is too large, it says so and gives both numbers. Shorten it yourself and run the same command again, ONCE. Do not loop: if the second attempt is refused as well, show the user the error and stop. The bridge will not shorten it for you, deliberately, because cutting from the front keeps your opening and drops your reasoning.
  6. If the command fails AND its error output mentions --adopt, an unlinked session was discovered heuristically: relay the error's timestamps to the user (never IDs), ask them to confirm it is the right session, and ONLY after an explicit yes rerun the exact same command with --adopt appended. For any other failure, show the user the exact error output and stop. Do not improvise other workarounds.
  7. If it succeeds, reply with exactly one short sentence like "Handing off to Claude — the bridge will switch automatically." and END YOUR TURN immediately. Do not run any further commands. The bridge launcher detects the completed turn and performs the switch; never mention session or thread IDs.

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.