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 ledq/resumery --skill bank-mapgit clone --depth 1 https://github.com/ledq/resumeryWrote 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/ledq/resumery/bank-map)<a href="https://agentmods.dev/skills/ledq/resumery/bank-map"><img src="https://agentmods.dev/badge/skills/ledq/resumery/bank-map.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 | $0.00054 | $0.00738 |
| Opus 5 | $0.00027 | $0.00369 |
| Sonnet 5 | $0.00011 | $0.00148 |
| Haiku 4.5 | $0.00005 | $0.00074 |
Grade A, and why
bank-map 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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/bank-map: regenerate the evidence map
bank/evidence_map.md is a derived index of bank/experience_bank.md, read by the
draft stage to turn its JD-coverage pass into a lookup instead of a per-run re-derivation.
It is never a source of truth: every resume claim traces to the bank, and on any
disagreement the map is stale and the bank wins. Regenerate it whole: read the entire
bank fresh and rewrite the file; do not patch the old map.
Structure (three sections, in this order)
- Requirement lookup: what JDs ask → strongest bank evidence, grouped under theme
headings (platform/backend, full-stack/frontend, data engineering, cloud/DevOps,
databases/performance, ML/AI, systems/embedded, collaboration/communication; adjust
themes to what the bank actually contains). Each line: the requirement as JDs phrase
it → the evidence, tagged with
[role_id], plus the headline facts and a production/ deployed marker where true. - Angle inventory: one entry per bank unit (each contribution and project). Header
carries
[role_id], ownership (sole/partial), and status (production, POC, shipped, competition). Body lists the distinct honest facings the unit supports (e.g. a config platform may face schema design, API design, distributed systems, and IaC), each angle with the exact facts (numbers, tools, tradeoffs, outcomes) behind it. - Absences and standing adjacency candidates: bank-wide absence facts (languages, clouds, platforms, role types the bank does NOT evidence), verified by scanning the whole bank, so the drafter can settle gap verdicts without a full re-scan; then the nearest-honest-evidence candidates for requirements that recur in postings, marked explicitly as per-JD decisions.
Rules
- Facts, never prose. Telegraphic fragments only; no sentence that could be lifted onto a resume. If an entry reads like a finished bullet, break it back into facts.
- Numbers verbatim from the bank. Metrics, costs, scale, dates: copy exactly; never round further or extrapolate.
- Every entry keyed.
[role_id]on everything, so the drafter maps evidence to resume structure without re-reading headers. - Angles must be honest facings, not aspirations. An angle exists only if the bank's facts under it would survive the evaluator's grounding check on their own.
- Absence claims are claims. State an absence only after checking the entire bank, including the projects and skills sections.
- Never launder placeholders. Bracketed TODOs or unfilled fields in the bank are not facts; where a unit has them, note the boundary ("do not claim beyond the above").
- Stamp it. Open the file with the generation date, the regeneration command, and the derived-not-source contract.
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 · 52 lines · 54 tokens per session scan A 694e7bec221c
bank-map is a skill published in the GitHub repository ledq/resumery (1 stars, last pushed 1mo ago), licensed MIT. It adds 54 tokens to every session and 738 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
seo
Use and read this skill immediately if the user request is in any way related to SEO or a site's organic search or AI search presence. That includes site audits, rankings, keyword research, competitors, backlinks, click or traffic changes, indexing problems, crawling, redirects, sitemaps, metadata, structured data…
pad
Talk to your project. Natural-language project management — create items, check status, plan work, brainstorm ideas, and more.
gno
Search local documents, files, notes, and knowledge bases. Index directories, search with BM25/vector/hybrid, get AI answers with citations. Use when user wants to search files, find documents, query notes, look up information in local folders, index a directory, set up document search, build a knowledge base, needs…
capture
Capture a thought as a Pad item without breaking flow.
connect
Connect this session to Pad's push/watch stream — consent to receive pushes, and run the workspace's on-session-start ritual on first connect.
memorywhale
Query and write durable debugging memory recorded by MemoryWhale. Use when debugging a failure that may have happened before, when you need the exact error/flags/output from an earlier attempt, when the user asks "how did we fix this last time?", or once you've figured out why something failed / how a fix worked and…