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 sungurerdim/dev-skills --skill ds-freezegit clone --depth 1 https://github.com/sungurerdim/dev-skillsWrote 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/sungurerdim/dev-skills/ds-freeze)<a href="https://agentmods.dev/skills/sungurerdim/dev-skills/ds-freeze"><img src="https://agentmods.dev/badge/skills/sungurerdim/dev-skills/ds-freeze/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/sungurerdim/dev-skills/ds-freeze"><img src="https://agentmods.dev/badge/skills/sungurerdim/dev-skills/ds-freeze.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00055 | $0.04542 |
| Opus 5 | $0.00028 | $0.02271 |
| Sonnet 5 | $0.00011 | $0.00908 |
| Haiku 4.5 | $0.00006 | $0.00454 |
Grade A, and why
ds-freeze 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 6d 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ds-freeze
Every feature stays "in scope" until someone explicitly says otherwise, so a release grows until every corner needs polishing and nothing ships — perfecting every detail is a losing race against scope creep. This skill forces the corner call: inventory every candidate feature and open issue, decide ship / defer-hidden / defer-backlog for each with the user, then only touch the kept set — implement it, file the rest as tracked backlog, and sync docs so nothing overclaims.
Release scope triage — collaborative feature-freeze, GitHub-issue-backed backlog, kept-set implementation, doc sync.
Completion Evidence — applies to every phase: Report
done/OKonly with the machine-checkable evidence the gates name — the exact command run and its observed output (orfile:linediff). Missing evidence → reportINCOMPLETEplus what is missing. Self-assessment is never evidence. (This band repeats at file end by design — both copies are normative.)
Triggers
- User asks to simplify or cut down a release, define an MVP feature set, decide what ships now vs later, or review open issues together to set release scope.
- Invoked by
/ds-shipvia the Scope-Freeze branch when the user's ask signals scope reduction before a launch-gate cascade.
Triggers — INVOKE / DON'T INVOKE
| INVOKE | DON'T INVOKE |
|---|---|
| "simplify this release, cut non-critical features" / "too much to perfect every detail before release" | "audit if we're ready to ship" (no scope-cut ask → ds-ship) |
| "decide what ships in v1 vs later" / "define the MVP feature set together" | "remove dead code / simplify the architecture" (→ ds-simplify) |
| "review our open issues and figure out release scope together" | "file this one issue" (single item, no triage → ds-issue) |
| "freeze scope before we harden everything for launch" | "what's the ideal architecture vs competitors" (→ ds-benchmark) |
Contract
Dimensions: none (carrier)
- One skill for the whole scope-freeze loop: candidate list → three-way disposition (
ship/defer-hidden/defer-backlog) → release manifest → implement the kept set → sync docs. - Never deletes code.
defer-hiddengates a feature behind a flag/toggle so it is unreachable this release; permanent removal is ds-simplify's job, with its own approval gate — ds-freeze only proposes it as a follow-up. - Undecided items default to
defer-backlog— never silently ships an item nobody confirmed (W5). - State-exempt. The tracking artifact (a GitHub issue, or
docs/release/{milestone}-scope.mdwhenghis unavailable) plus git are the durable record; nothing is written tods/audit/. Resuming = re-read the tracking artifact and its comments — decisions land in the thread as often as the checklist. - Standalone — advisory handoff to ds-blueprint, ds-build, ds-issue, ds-docs, ds-backend/ds-frontend/ds-review, ds-simplify, ds-test (scope per skill: see Delegation); absent → inline fallback or gap-note, run continues either way. Every delegated call forwards no flags by default;
--askpropagates to it. - Pre-existing / out-of-scope errors detected during work are NOT skipped — fixed inline or escalated with concrete blocker.
- Full accounting enforced: every finding and planned check ends in an explicit disposition (fixed / skipped + reason / only you can do); summary totals balance.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago Changed · -7 lines · -4 tokens per session e8e40bacca18
- 9d ago First seen · 160 lines · 59 tokens per session scan A eacddba41418
ds-freeze is a skill published in the GitHub repository sungurerdim/dev-skills (1 stars, last pushed 4d ago), licensed MIT. It adds 55 tokens to every session and 4,542 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
execution/release-executor
Internal: Release Executor. Implements version/changelog updates, rollout flags, rollback plans, release notes, and migration sequencing. Write evidence on completion.
add-changelog-entry
Add a changelog entry to UNRELEASED.md.
memstack-development-changelog-generator
Use when the user says 'generate changelog', 'update changelog', 'what changed', 'release notes', 'write changelog', or needs a formatted CHANGELOG.md from git commit history. Do NOT use for diary entries, git log viewing, or commit message writing.
ijfw-ship
Use when the user says 'ship it', 'ship this', 'release', 'publish', 'launch', 'deploy', 'go live', 'wrap this up', 'time to ship', or invokes '/ijfw-ship'. Domain-aware release — software (test → tag → push → publish), book (final edit → format → KDP / agent / Substack), campaign (review → schedule → launch →…
ijfw-preflight
Run the IJFW preflight pipeline (11 gates, fail-fast). Trigger: 'ijfw preflight', 'run preflight', 'check before ship', 'preflight gates', 'validate before release'.
ring:committing-changes
Commit changes with scope allowlist enforcement, atomic grouping, GPG-signed conventional commits, and trailer management. Detects the repo's PR-validation scope policy before proposing any message. Use when the user asks to commit or has changes ready to record. Skip when the working tree is clean or the user wants…