Agent skill · Audited from source
Yc Apply
skill · pedronauck/skillsLicence not declaredAudited 16 Oct 2026 · source 118e61cf1c61
Is it safe? Yes — SAFE, grade B (89/100). We read this skill's source at commit 118e61cf1c61 and found no filesystem, network or shell access and no hard-coded secrets. What it does: Prepare, review, and package Y Combinator application answers and founder-video notes using the live form and verified founder facts. Use for a YC application, requested interview practice, or reapplication; not general…
Cite
Cite this page
ModelsAtlas. “Yc Apply (pedronauck/skills)”. Retrieved 16 Oct 2026. https://modelsatlas.com/skills/pedronauck/yc-apply
Trust audit
Audit 0.4.1 · source read 16 Oct 2026
SAFE means nothing in the source contradicts what the skill says it does. Grade B is shared by 16,667 of 17,602 audited skills.
What the source does
- FilesystemNone observedNo reads or writes in the source
- NetworkNone observedNo outbound calls in the source
- ShellNone observedNo commands run in the source
- DependenciesPinnedNothing floats
- SecretsNone foundNo keys in source
Five audit layers
| Layer | What we check | Result | Findings |
|---|---|---|---|
L0 Provenance | Repo owner, licence, archive status, activity | PASS | 0 |
L1 Static source | Secrets, path traversal, supply chain, hidden code | PASS | 0 |
L2 Instructions | Prompt injection, authority framing, hidden directives | PASS | 0 |
L3 Inventory | The files and declarations match the source | PASS | 0 |
L4 Sandbox | Run in isolation and watch behaviour | SKIPPED why | 0 |
Gates applied: no behavioural pass.
Findings (0)
No findings at any severity. We re-audit when the source changes.
Audit 0.4.1 · source sha 118e61cf1c61 · 16 Oct 2026Report an issue or request a re-scan
Provenance
Who published it, and what the verdict is pinned to
- Publisherpedronauck GitHub userOwner of the repository this skill is read from
- Source folderpedronauck/skills/skills/mine/yc-applyThe exact folder we audited
- LicenceNot declaredAs the repository states it
- Commit pinned for this verdict118e61cf1c61A new audit runs when the source changes
- Repository statusNot archivedLast push 31 days ago
- Sandboxed runNot runSkills are read, not executed; layer L4 is skipped
What it tells the agent
Full source on GitHub ↗Its description: Prepare, review, and package Y Combinator application answers and founder-video notes using the live form and verified founder facts. Use for a YC application, requested interview practice, or reapplication; not general…
--- name: yc-apply description: Prepare, review, and package Y Combinator application answers and founder-video notes using the live form and verified founder facts. Use for a YC application, requested interview practice, or reapplication; not general fundraising or unrelated decks. disable-model-invocation: true --- # YC Apply Complete the requested application task using the founder's facts and the current YC form. Work can start with one answer, a review, video notes, a submission pack, or reapplication; it need not traverse every coaching phase. ## Workspace and context Keep founder data in the separate workspace they chose. Resolve `--workspace`/`--resume`, `YC_APPLY_WORKSPACE`, or an existing `00_meta.md`; ask only when no unambiguous location is known. `<skill-dir>` is this skill's directory, not the workspace. Create the authorized workspace with `scripts/bootstrap-workspace.sh`; preserve existing files and resume instead of overwriting them. Use the existing layout: `00_meta.md`, `01_form-spec.md`, founder/narrative notes, optional `04_research/`, `05_drafts/`, `06_video/`, `06_gate.md`, optional `07_interview/`, `08_final/`, and milestone journal. These are artifact locations, not prerequisites for every task. Legacy `--phase` numbers may select the corresponding task; they no longer impose a sequential gate. Capture/verify the live form in `01_form-spec.md` for a full application or submission check. Bundled batch forms are dated snapshots, not current authority. A single-answer edit can use the supplied field and limit without a full form-research pass. ## Choose the work - **Draft or improve answers:** reuse known facts, batch missing factual questions, and ask a focused follow-up for a consequential ambiguity. Start from a concrete draft when the founder requested drafting. Use matching `references/partner-signals.md`, `references/seven-pitch-questions.md`, and `references/accepted-examples.md` sections when helpful; no compulsory seven-question interview, score threshold, Email Test, or accepted-example quota. - **Research:** investigate unresolved claims that affect the answer. Use applicable `assets/research-*.md` outlines and bounded native explorers only when independent questions justify them. No five-slice prerequisite. Record evidence and uncertainty; a legitimately inapplicable regulatory slice may be `N/A` without an invented URL. - **Review:** apply the relevant `references/anti-patterns.md` guidance. `scripts/buzzword-scan.sh` suggests clarity candidates; a technical word is not itself a defect. No journal entry is required for every intentional word. Preserve truthful meaning, exact field limits, and the founder's voice. - **Founder video:** read `references/video-spec.md` and verify current application instructions. Preserve its time/participant/content requirements; use concise talking notes by default. An explicit request for a script is sufficient authorization to draft one; explain the natural-delivery tradeoff without requiring a magic confirmation phrase. - **Interview:** use `references/interview-playbook.md` for requested practice. Mark the actual interview stage only after a confirmed invite; practice does not imply an invitation. Choose drills for weak answers, with optional timers; no personality performance or fixed rehearsal quota. - **Reapplication/post-mortem:** use `references/reapplicant-playbook.md` for confirmed changes since the prior application. Do not auto-launch a rejection interview or scaffold another batch unless requested. ## Packaging and submission readiness For a requested final pack, collect the accepted answers in `08_final/SUBMIT.md` in current form order. Record the founder's review and the material readiness checks in `06_gate.md`: ```text form_current: pass required_fields_complete: pass claims_verified: pass video_conforms: pass founder_reviewed: pass ``` Use `video_conforms: not_required` only when the current form truly requires no video. A missing/failed item remains an explicit gap; optional research, coaching, and scoring cannot block otherwise ready answers. `scripts/phase-status.sh <workspace> --submission` checks the record and artifacts; it cannot independently prove the truth of a founder claim or the live form's requirements. The pack is for founder review/use. Actual submission, emails to others, uploads, or disclosure of transcripts require explicit authorization. Preserve existing authorization without asking twice. Never invent metrics, customers, evidence, founder consent, or successful submission. Inspect a selected coding transcript for private data and current upload limits before staging it; do not explore unrelated conversation history. Use `scripts/journal-append.sh` for meaningful decisions/handoffs, not every answer or phase. `scripts/phase-status.sh <workspace>` is an advisory progress inventory, not an instruction to restart completed work. An optional `--strict` buzzword scan may support a specifically requested lexical audit; it is never proof of application quality.
Excerpt, first 5,067 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.
Install
Pinned to the audited commit 118e61c
git clone https://github.com/pedronauck/skills git -C skills checkout 118e61cf1c61
The author documents no install command, so this uses the repository directly. Our verdict covers commit 118e61cf1c61 only.
Works in
The author has not said which agents this skill supports, and we have not tested it in any.
Audit history
Every audit, every source change
- ✓SAFE B 89 / 100
first audit · source
118e61cf1c61
Similar skills
obra/superpowers
Use when creating new skills, editing existing skills, or verifying skills work before deployment
mattpocock/skills
A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.
mattpocock/skills
Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.
Questions about Yc Apply
What does the Yc Apply skill do?
Is Yc Apply safe to install?
What can Yc Apply access on my machine?
How current is this page?
Sources
- github.com/pedronauck/skills/tree/main/skills/mine/yc-apply — the audited source at commit 118e61cf1c61 and every finding aboveObserved · 16 Oct 2026
Spotted something wrong? Report a correction · Methodology
Keep going in your AI
Opens with this page as the source. We add nothing else to the prompt.
- What does the Yc Apply skill do?
- Is Yc Apply safe to install?
- What can Yc Apply access on my machine?