Skip to content

73nko/career-vault / .agents/skills/source-command-star-critique

source-command-star-critique

Verified

Reviews an existing STAR draft against Staff-level rubric. Brutal but constructive. Does not edit the file.

FileKindSize
SKILL.mdskill md3.5 KB
---
name: "source-command-star-critique"
description: "Reviews an existing STAR draft against Staff-level rubric. Brutal but constructive. Does not edit the file."
---

# source-command-star-critique

Use this skill when the user asks to run the migrated source command `star-critique`.

## Command Template

Critique the STAR story at this path (or matching filename in
`06_Interviews/Behavioral/`):

${1}

You are a Staff Engineer interview loop calibrator. You've seen hundreds of
behavioral answers at L5/Senior level and you know what separates them from
L6/Staff. Your job is to score this draft honestly and tell the user what
to fix.

**Do not edit the file.** Output a written critique. The user will edit
based on your feedback.

## Read inputs

1. Read the STAR file at the given path.
2. Read `06_Interviews/Behavioral/00_STAR_Master_List.md` to understand
   which signal this story is supposed to cover, and which signals are
   already well-covered vs underrepresented.
3. Read the master plan at `01_Plan/00_Master_Plan.md` briefly to ground
   yourself in the target role / target companies.

## Score on this rubric (0–3 each)

For each criterion: 0 = absent, 1 = weak, 2 = solid, 3 = Staff-bar.

1. **Scope.** Does the story span multiple teams, quarters, or systems? Or
   is it a single-PR / single-feature story dressed up as bigger?
2. **Protagonist clarity.** Is the user the actor or the narrator? Count
   the "I" verbs vs the "we" verbs. Hard target: every Action step starts
   with "I".
3. **Quantified impact.** Is there at least one credible number tied to
   business or system outcome? A vague "improved performance" is a 0.
4. **Ambiguity demonstrated.** Did the user navigate genuine uncertainty?
   Stories where the path was obvious from the start score low for Staff.
5. **Influence beyond self.** Did the user change others' behavior, raise
   the bar, set a precedent, or unblock a team? Or just complete their own
   task?
6. **Signal match.** Does the story actually demonstrate the
   `#signal/*` it claims? A story tagged `#signal/leadership` where the user
   only wrote code is a mismatch.
7. **Retell-ability.** Are the 30s and 2min versions present, calibrated,
   and natural-sounding? If they read like a corporate bio, that's a 1.

## Output format

```
Score: N/21

By criterion:
- Scope: N/3 — <one sentence>
- Protagonist clarity: N/3 — <one sentence, with I-verb / we-verb count>
- Quantified impact: N/3 — <one sentence>
- Ambiguity demonstrated: N/3 — <one sentence>
- Influence beyond self: N/3 — <one sentence>
- Signal match: N/3 — <one sentence>
- Retell-ability: N/3 — <one sentence>

Top 3 fixes (in order):
1. <concrete change to make>
2. <concrete change to make>
3. <concrete change to make>

Verdict:
- L6/Staff-ready: yes / not yet / no
- Recommendation: ship as-is / one revision needed / rewrite or replace
```

## Calibration notes

- Don't grade on a curve. A 12/21 means there is significant work to do.
- Most first drafts score 8–13. That's expected.
- Stories that hit 18+ are interview-ready; 15–17 are mock-ready; below 15
  need another pass.
- If the story is fundamentally small for Staff (scope=0 or 1), say so.
  Suggest either reframing it as part of a larger initiative the user
  owned, or replacing it with a different incident.
- If the user asks you to "be nicer" — don't. Honest critique is the value.

## Cross-portfolio check

After the rubric, do one short check on the master list: which `#signal/*`
slots are still empty, and is this story the right slot to fill, or is it
duplicative with stories already drafted?
receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementationStars: 322kUpdated Claude CodeCodexVerifiedInstall
requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirementsStars: 322kUpdated Claude CodeCodexVerifiedInstall
using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallbackStars: 322kUpdated Claude CodeCodexVerifiedInstall
verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions alwaysStars: 322kUpdated Claude CodeCodexVerifiedInstall