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 CrumbtrailDev/crumbtrail-cli --skill flag-landminegit clone --depth 1 https://github.com/CrumbtrailDev/crumbtrail-cliWrote 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/crumbtraildev/crumbtrail-cli/flag-landmine)<a href="https://agentmods.dev/skills/crumbtraildev/crumbtrail-cli/flag-landmine"><img src="https://agentmods.dev/badge/skills/crumbtraildev/crumbtrail-cli/flag-landmine.svg" alt="Measured on agentmods" 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 | $0.00071 | $0.01548 |
| Opus 5 | $0.00036 | $0.00774 |
| Sonnet 5 | $0.00014 | $0.00310 |
| Haiku 4.5 | $0.00007 | $0.00155 |
Grade A, and why
flag-landmine 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 7d 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The flag and toggle landmine
Flags accumulate three jobs at once: configuration, rollout control, and an archive of past workarounds. They are usually free text and usually unvalidated. A toggle nobody has touched for years blocks a new code path, or a flag that was never initialised for some users silently disables a feature for exactly them.
Before running this skill, confirm the recording actually carries a flags map. If it does not, this archetype is invisible to the recording and the first useful action is to make the application declare its flags.
Symptom
Two users on the same build get different behaviour. A feature is off for some accounts and nobody can name the switch. A new path works in one account and short circuits in another. The reporter often frames this as intermittent, because from their side the same action produces different results, but it is stable per user rather than random.
What Crumbtrail can see
The environment snapshot emitted at session start carries the feature flags and the runtime
configuration the application declared to the SDK, redacted, and later declarations arrive as
deltas on the same channel. getRegressionContext compares two recordings and returns an
environment delta naming each flag key, config key, release label and build label that moved
between them, so a working session and a broken session can be diffed directly.
Alongside that, the session carries the usual planes: interactions, console, requests and responses, backend spans where the backend integration is installed, and database row changes where the database adapter is installed. Those are what tell you the flag actually changed a code path rather than merely differing.
It does not read your flag provider. Crumbtrail has no connection to a flag service and cannot enumerate the flags that exist, their defaults, their targeting rules, or their history. Only the values the application chose to declare are present, and only from the moment it declared them. If the application never declares its environment, the flags section is absent and no query in this skill will produce anything.
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.
- 7d ago First seen · 149 lines · 71 tokens per session scan A 3d4ac816ebe2
flag-landmine is a skill published in the GitHub repository CrumbtrailDev/crumbtrail-cli (0 stars, last pushed today), licensed MIT. It adds 71 tokens to every session and 1,548 once invoked, about $0.0004 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
cce-routing
Use Cursor Bridge's read-only cursorcontextengine for unfamiliar project understanding when the exact code location is unknown or the task requires tracing behavior, symbols, callers and callees, data flow, registrations, interface implementations, ownership boundaries, or cross-module relationships. Trigger for…
gh-fix-ci
Use when a user asks to debug or fix failing GitHub PR checks that run in GitHub Actions; use gh to inspect checks and logs, summarize failure context, draft a fix plan, and implement only after explicit approval. Treat external providers (for example Buildkite) as out of scope and report only the details URL.
sentry
Query Sentry organizations, projects, and unresolved issues/errors via the Sentry API. Use when the user asks about Sentry errors, issues, or projects.
signoz
Query observability (services, traces, logs, metrics) from a SigNoz instance via its API. Use when the user asks about app performance, errors, latency, or telemetry in SigNoz.
rel-ai-debugging
Use when repository behavior is reproducibly wrong and needs causal diagnosis or repair, including errors, broken tests, crashes, regressions, failed contracts, or incorrect runtime behavior. Do not use for general audits or final verification when no active defect is being diagnosed.
rel-ai-investigation
Use for read-only repository questions that need evidence, including architecture audits, feasibility studies, dependency or caller tracing, impact analysis, implementation-status checks, and proof of how something works. Do not use to implement fixes or for final completion or release verification of changes already…