simiancraft/simiancraft-skills / skills/appraise-github-issues
appraise-github-issues
VerifiedAppraise a GitHub issue backlog unattended: for each issue, decide whether it is still real and how big it is, then apply the answer to the tracker as a size label, a needs-decision or needs-human label with the question stated, or a close carrying a re-checkable receipt. A close is confirmed by a second engine before it is made. Runs once, on one issue, over the whole backlog, or as a periodic heartbeat, and is the sizing stage of burn-down-github-issues. Use when the task is "size the backlog", "appraise every issue in this repo", "close what is already done", "triage recent issues without fixing them", or "keep the backlog sized". Requires a per-repository config (its own, or the burndown's) and the codex and claude CLIs for the default seats. Skip for fixing anything; this skill never writes to the repository.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.1 KB |
| appraise.ts | script | 12.7 KB |
| lib/appraise.test.ts | script | 2.8 KB |
| lib/appraise.ts | script | 30.9 KB |
| lib/callbacks.ts | script | 4.1 KB |
| prompts/appraise.md | markdown | 6.2 KB |
| prompts/confirm-close.md | markdown | 3.3 KB |
| references/adopting.md | markdown | 4.0 KB |
| references/callbacks.md | markdown | 3.4 KB |
references/callbacks.md
View on GitHub# Size callbacks
What happens after an issue is sized is not the appraiser's decision. A producer (the issue
burndown, a person, another skill) puts files into a directory, and the appraiser looks one up for
the points it just applied and runs it, knowing nothing about what it does. This is the same slot
mechanism the burndown's floor uses for `on-walk` and `on-walk-fail`, keyed by size instead of by
walk.
## The directory
`callbacksDir` in the config (default `<worktreeRoot>/appraisal-callbacks`, relative to the
repository root), or `--callbacks <dir>` for one run. The burndown writes the callbacks it ships
into that directory on every start, overwriting its own file names and leaving any others alone,
so an adopter can add a local callback beside the shipped ones.
## The ladder
A Fibonacci scale does not need a file per rung. For an issue sized at N points, the first slot
that exists wins:
| Slot | Matches |
|---|---|
| `on-size-<N>` | exactly N |
| `on-size-over-<M>` | the file with the largest M below N |
| `on-size` | any size |
So `on-size-over-8.md` and `on-size-over-21.md` together mean: 13 runs the first, 34 and up run the
second, and nothing runs for 8 or less unless `on-size` exists.
## The two forms
Each slot name may exist as an executable, as a Markdown prompt (`<name>.md`), or both.
- **The executable** runs first, with the appraisal payload as one line of JSON on stdin, its own
process group, a timeout of `sizeCallbackTimeoutMinutes` (default 0: no timer, since a callback
that runs agents cannot be bounded by a minute), and its exit code logged. This is the form for anything load-bearing: a
label, a pause, a notification. An executable never depends on an agent following prose.
- **The prompt** is one agent turn on the callback seat (`seats.callback`, by default the
appraiser's own seat), in a scratch directory that is not a checkout. It is rendered with the
project vocabulary plus `{{ISSUE}}`, `{{TITLE}}`, `{{POINTS}}`, `{{PRIOR_POINTS}}`,
`{{REASON}}`, and `{{CALLBACK_FILE}}`. It may load another skill by name; that is how a producer
reaches for work the appraiser does not do, such as breaking a large issue into smaller ones.
The payload on stdin:
```json
{ "issue": 1234, "title": "…", "points": 13, "priorPoints": null, "verdict": "valid",
"reason": "…", "repo": "owner/repo", "baseBranch": "main", "repoRoot": "/abs/path/to/checkout" }
```
## What a prompt may report
A prompt that writes `loop-callback.json` in its scratch directory has its outcome recorded in the
log and returned to the caller:
```json
{ "outcome": "split", "reason": "three sub-issues opened; see the parent's thread" }
```
Both fields are free text. A prompt that writes nothing is logged as finished without an outcome.
Nothing a callback does can fail the appraisal: the size is already on the issue by the time the
callback runs, and a callback that exits non-zero or crashes is logged and the appraisal continues.
## Rules the appraiser keeps
- A callback runs only for `valid` verdicts, after the size is on the issue. Closes and hand-offs
have no callback; a close is confirmed instead, and a hand-off is already a person's.
- A callback prompt runs on a seat with the same trust as the appraiser: read-only against the
repository, able to write to the tracker. Whatever it writes to the tracker is the producer's
responsibility, and an issue it creates is appraised on a later pass like any other.
- `--dry-run` runs neither form.
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: 471k | 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 |