Borrowing it
Nothing to install: this file belongs to timfewi/tentaflake. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/timfewi/tentaflake/main/.agents/skills/tentaflake-repo-guidance/SKILL.mdgit clone --depth 1 https://github.com/timfewi/tentaflakeWrote 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/timfewi/tentaflake/tentaflake-repo-guidance)<a href="https://agentmods.dev/skills/timfewi/tentaflake/tentaflake-repo-guidance"><img src="https://agentmods.dev/badge/skills/timfewi/tentaflake/tentaflake-repo-guidance/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/timfewi/tentaflake/tentaflake-repo-guidance"><img src="https://agentmods.dev/badge/skills/timfewi/tentaflake/tentaflake-repo-guidance.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 | $0.00021 | $0.00767 |
| Opus 5.5 | $0.00008 | $0.00307 |
| Sonnet 5.5 | $0.00004 | $0.00153 |
| Haiku 4.5 | $0.00002 | $0.00077 |
Grade A, and why
tentaflake-repo-guidance 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 3d 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.
What it actually says
Repository guidance
Read AGENTS.md and CONTRIBUTING.md, then inspect only the relevant owners:
| Area | Source | Guide |
|---|---|---|
| Runtime definition | adapters/<id>.nix, lib/mkAgent.nix, lib/runtimeContract.nix |
Adapters |
| Onboarding/inventory | lib/systemFromData.nix, lib/agentPlan.nix, CLI |
Configuration |
| Containment | lib/containerSecurity.nix, modules/security.nix |
Profiles |
| Host services | modules/{broker,research,worker,workspace-quota,backup}.nix |
Operations and linked policy guides |
| Management | modules/management.nix, modules/tailscale.nix |
Management |
| CLI/API | crates/tentaflake-cli/, host-api/ |
CLI, API |
| Installer/builds | installer/disk.sh, lib/mkRustPackage.nix, scripts/eval_flake.py |
Install, Builds |
Define instances only in tentaflake.toml; public systemFromData accepts only
{ file = ./tentaflake.toml; }. Keep the public inventory empty and commented.
Load real TOML from the consumer/fork's tracked Git source, never an upstream
template fallback or path:. filtering bypass; see the configuration guide.
Refuse legacy agents.toml/agents.json and missing/Git-filtered desired sources.
mkAgent and adapter builds are internal. Keep
native behavior in the selected adapter and host authority in Nix policy,
which may consume metadata but must not define more instances.
Public TOML accepts only descriptor-defined fields; internal adapter arguments
such as settings and seedDir are not a second public configuration path.
Adapter preset facts are the editable source;
python3 scripts/runtime-catalog-docs.py regenerates the JSON/docs projections.
Names are globally unique and host IDs are agent-<name>. Use metadata for
paths/owners; never move existing data or reset budgets implicitly.
Preserve balanced startup gates, host-held secrets, exact read-only Research mounts, offline workers and quota/backup ownership. No native tool mediation or vendor acceptance follows from a declaration or synthetic fixture. Scaffolds stay non-runnable without OCI/state/credentials. Generic custom OCI grants no accepted model/Research integration. For authority changes read the threat model.
Choose checks through the build guide: just fast includes pure catalog drift
and documentation inventory checks; local host evaluation/build/runtime evidence
remains separate, while CI runs every affected gate. Update owning guides and
examples with behavior, keep skills short, and report actual checks and
remaining evidence gaps. Evaluation does not authorize activation.
docs/ is product documentation only (see its guide inventory). Write plans,
notes, prompts and investigation results to the ignored .scratch/; never
commit them or private deployment facts.
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.
- 3d ago Changed · +2 lines 242ac31e3ff4
- 4d ago Changed · -53 lines · -8 tokens per session 1f20c39cd19d
- 17d ago Changed · -21 lines 0fd97b4a38e5
- 23d ago Changed · +28 lines 3bda8e9a7bd3
- 1mo ago Changed · +12 lines 7d7b0d661ee9
- 1mo ago Changed · +22 lines 9664f9afab50
- 1mo ago Changed · -427 lines · -3 tokens per session 8dee319f3431
- 1mo ago Changed · +27 lines 4591181648b9
- 1mo ago Changed · +79 lines 293523024dad
- 2mo ago First seen · 383 lines · 32 tokens per session scan A 8eb6726cc28c
tentaflake-repo-guidance is a skill published in the GitHub repository timfewi/tentaflake (32 stars, last pushed yesterday), licensed MIT. It adds 21 tokens to every session and 767 once invoked, about $0.0001 per session on Opus 5.5. 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.
tentaflake-repo-guidance compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| tentaflake-repo-guidancetimfewi/tentaflake | Skill | 32 | yesterday | MIT | A | 21 | Claude Code, Codex |
| memory-safe-migrationcosai-oasis/project-codeguard | Skill | 364 | 10d ago | none | A | 117 | Claude Code |
| fix-security-vulnerabilities-with-strixusestrix/strix | Skill | 74k | yesterday | Apache-2.0 | A | 124 | Claude Code, Codex |
| python-feature-lifecyclemicrosoft/agent-framework | Skill | 15k | yesterday | MIT | A | 43 | Claude Code, Codex |
| effect-v4ComposioHQ/composio | Skill | 31k | yesterday | MIT | A | 159 | Claude Code, Codex |
| ci-security-scanning-with-strixusestrix/strix | Skill | 74k | yesterday | Apache-2.0 | B | 140 | Claude Code, Codex |
| python-developmentmicrosoft/agent-framework | Skill | 15k | yesterday | MIT | A | 27 | Claude Code, Codex |