dongshuyan/compass-skills / skills/run-history-skill-upgrader
run-history-skill-upgrader
VerifiedUse real run evidence, validation failures, source drift, platform drift, and user feedback to plan and, only after explicit approval, apply structural upgrades to an existing skill. Use when the user asks to improve an existing skill from recent runs, recurring failures, outdated sources, excessive bloat, changed platform behavior, or validated workflow feedback. Do not use to create a brand-new skill or to execute the business workflow itself.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 6.6 KB |
| agents/openai.yaml | config | 306 B |
| evals/evals.json | config | 5.4 KB |
| references/evidence-and-scope.md | markdown | 2.6 KB |
| references/examples.md | markdown | 2.3 KB |
| references/upgrade-decision-protocol.md | markdown | 2.9 KB |
| references/validation-and-regression.md | markdown | 2.7 KB |
| scripts/test_validate_upgrade_artifacts.py | script | 5.7 KB |
| scripts/validate_upgrade_artifacts.py | script | 13.6 KB |
references/examples.md
View on GitHub# Examples
These are public, sanitized upgrade patterns. They are not a license to turn every incident into a permanent rule.
## Example 1: Missing Finalization Gate
Signal:
- a workflow claimed success before the final processed artifact actually existed.
Good upgrade:
- add or strengthen a finalization gate;
- move the success claim behind a deterministic check;
- update examples and validation accordingly.
## Example 2: Source Drift
Signal:
- a previously used source now looks inaccessible.
Good upgrade:
- check whether the problem is temporary, authentication-related, moved, rate-limited, or truly discontinued;
- replace or deprecate only with strong evidence;
- keep the plan conservative if evidence is weak.
## Example 3: Durable vs Temporary Preference
Signal:
- the user wants a certain output style or field included every time.
Good upgrade:
- keep it durable only if the user clearly wants the preference preserved across future runs;
- otherwise treat it as task-local and avoid changing the skill.
## Example 4: Candidate Idea
Signal:
- an agent or user suggests parallelizing the skill.
Good upgrade:
- keep it as a candidate idea until run evidence shows that concurrency solves a real bottleneck without unsafe shared-state collisions.
## Example 5: Prune Or Merge
Signal:
- the skill keeps growing and repeats the same rule in multiple places.
Good upgrade:
- identify the strongest structure owner;
- merge duplicate prose;
- delete low-value reminders that add no validation power.
## Example 6: Platform Drift
Signal:
- a familiar UI changed enough that the old workflow no longer lines up with visible states.
Good upgrade:
- update the documented visible checks and any affected selector or navigation logic;
- keep user-error cases separate from platform-drift cases.
## Example 7: Overfit Incident
Signal:
- one failed path or one temp filename is proposed as a permanent rule.
Good upgrade:
- abstract to the underlying validation gate;
- if no clean abstraction exists, choose `no_change` or `maintenance_note_only`.
## Example 8: Plan-Only Boundary
Signal:
- the user asks for direct edits before seeing a plan.
Good upgrade:
- produce the plan, say that no files were modified yet, and wait for explicit approval that points to the plan.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |