Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add kalyvask/winning-writing --skill feedback-rephrasergit clone --depth 1 https://github.com/kalyvask/winning-writingWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kalyvask/winning-writing/feedback-rephraser)<a href="https://agentmods.dev/skills/kalyvask/winning-writing/feedback-rephraser"><img src="https://agentmods.dev/badge/skills/kalyvask/winning-writing/feedback-rephraser/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kalyvask/winning-writing/feedback-rephraser"><img src="https://agentmods.dev/badge/skills/kalyvask/winning-writing/feedback-rephraser.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00149 | $0.01662 |
| Opus 5 | $0.00075 | $0.00831 |
| Sonnet 5 | $0.00030 | $0.00332 |
| Haiku 4.5 | $0.00015 | $0.00166 |
Grade A, and why
feedback-rephraser scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Feedback rephraser
Source: points/performance-review-rules.md (reframing kit) and points/core-rules.md (say what you like + what you would like).
Why this exists
Blunt feedback feels honest to the giver and burns trust with the receiver. The receiver hears "you are bad at X" even when the giver meant "the work needs Y." The fix is mechanical: lead with what you LIKE, then what you WOULD LIKE, anchored to the WORK, phrased as a QUESTION where possible.
The principle generalizes across every channel — annual review, Slack DM, 1:1 talking point, peer review comment, HR script. Same rewrite, same rules.
What to detect
Catch any sentence that does one or more of:
- Addresses a personal trait, not a behavior in the work ("you don't smile enough," "you're intellectually intimidating," "you're too blunt")
- Tells the person to stop something without naming what to do instead ("stop interrupting," "stop being late")
- Uses you are / you're too / you don't as the load-bearing verb
- Speculates about the person's emotional state ("you're burning out," "you don't care")
- Mentions personal context the writer hasn't been invited to discuss ("because of your divorce," "since the baby was born")
- Phrases a complaint as a declaration when a question would land softer ("this isn't working")
The five-move reframe
Every rewrite uses some combination of these moves:
- Open with what you LIKE — at least one named, specific thing.
- Re-aim at the WORK — not their personality, not their emotional state. The deliverable, the meeting, the client interaction.
- Convert "stop X" to "I'd love to see more Y" — frame the positive behavior, not the prohibition.
- Phrase as a question — "can we talk about how to...", "is it time to...", "worth thinking about whether..."
- Cut all psychoanalysis — never speculate about why. If you suspect a personal reason, that's a separate, private conversation.
Reframing table
The patterns that show up most often. Detect any of these constructions and rewrite them.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 102 lines · 149 tokens per session scan A 605c7eba861b
feedback-rephraser is a skill published in the GitHub repository kalyvask/winning-writing (14 stars, last pushed 6d ago), licensed MIT. It adds 149 tokens to every session and 1,662 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
Pair Programming
Collaboratively design, implement, review, and refine software features with a human developer using an iterative pair programming workflow. Prioritize understanding, explicit approval, small reviewable changes, and synchronized specifications.
spike-consumer-baseline
OI-3 spike harness — heavy consumer skill, BASELINE arm. Builds an incident post-mortem and sources the readability standard mid-workflow by reading a file directly, with NO Skill-tool call. Use only when explicitly invoked by the spike harness with a TRIALID and data path.
spike-consumer-adversarial
OI-3 spike harness — heavy consumer, ADVERSARIAL arm. Worst-case early-exit test: the mid-workflow Skill call has no continuation guardrail and the guidance skill ends with a final-sounding anchor. Use only when explicitly invoked by the spike harness with a TRIALID and data path.
chief-of-staff
The CEO's right hand — maintains validated project context, prepares task contracts, logs cross-department decisions, and runs the weekly review. Use when the user says "what's in flight", "resume the project", "log this decision", "weekly review", "turn this idea into a brief", or when multi-department work needs…
atelier-orchestrator
Skill routing and workflow orchestration. Selects Inline Plan or Spec-backed Plan, routes to the correct workflow skill, and manages transitions between phases. Use when starting any conversation or task to determine which planning mode and skill apply.
close-management
Runs the month-end close: generates a day-by-day runbook from D-2 to D+5 as CLOSE- .md with an owner matrix, dependency-ordered tasks (subledgers → accruals → reconciliations → flux → lock), numeric flux thresholds, a preparer/reviewer sign-off log, and close metrics such as days-to-close and reopened entries. Use…