using-tackle
Skill0 starsChanged 45 days agoMITView on GitHub
Use when deciding how much process a coding task needs, or when unsure which Tackle skill fits the work in front of you
In this repository13 other files it gives its agents.
- debugging-systematicallySkill
- delegating-to-subagentsSkill
- executing-plansSkill
- finishing-workSkill
- investigating-in-parallelSkill
- isolating-workspacesSkill
- planning-workSkill
- receiving-reviewSkill
- requesting-reviewSkill
- shaping-designsSkill
- testing-firstSkill
- verifying-before-claimingSkill
- writing-skillsSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is 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.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
When in doubt between two sizes, pick the smaller and step up if surprises appear.
- Very specific: line 34
When several skills fit, start with the one that decides the approach (shaping-designs, debugging-systematically), then the ones that carry …
≈ “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.
Files
Using Tackle
Tackle skills are tools, not a pipeline. Pick the ones that fit the task, skip the rest, and say briefly which you're using when it helps the user follow along.
Scale it
| Size | Looks like | Process |
| Trivial | Typo, rename, config value, one obvious line | Do it, then verify |
| Small | One coherent change, approach is clear | State intent in a sentence, test first where a harness exists, verify |
| Substantial | Several components, unclear approach, or costly to undo | Shape the design, plan, execute with checkpoints |
When in doubt between two sizes, pick the smaller and step up if surprises appear.
How
Precedence: the user's explicit instructions (including project instruction files such as CLAUDE.md, AGENTS.md or GEMINI.md) come first, then the harness and its system prompt, then Tackle. A skill never overrides the first two.
Platforms: skills describe tools by role ("dispatch a subagent", "track tasks"). references/platforms.md maps each role to Claude Code, Codex and Gemini CLI, with fallbacks for missing ones.
Skill map:
- Shaping an idea:
shaping-designs→planning-work - Doing the work:
executing-plans,delegating-to-subagents,investigating-in-parallel,isolating-workspaces - Quality:
testing-first,debugging-systematically,verifying-before-claiming - Review:
requesting-review,receiving-review - Wrapping up:
finishing-work - Extending Tackle:
writing-skills
When several skills fit, start with the one that decides the approach (shaping-designs, debugging-systematically), then the ones that carry it out.
Skip or shorten when
- The user says how they want to work. That wins over any sizing.
- The task is trivial. Don't reach for a skill at all.
Done means
You've picked a size and the matching skills, or consciously none.
Next
Whichever skill fits. Each one ends with suggestions; you and the user decide.
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.