Borrowing it
Nothing to install: this file belongs to Kemetra/Seshat-BI. 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/Kemetra/Seshat-BI/main/.claude/skills/approval-evidence-pack/SKILL.mdgit clone --depth 1 https://github.com/Kemetra/Seshat-BIWrote 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/kemetra/seshat-bi/approval-evidence-pack)<a href="https://agentmods.dev/skills/kemetra/seshat-bi/approval-evidence-pack"><img src="https://agentmods.dev/badge/skills/kemetra/seshat-bi/approval-evidence-pack/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/kemetra/seshat-bi/approval-evidence-pack"><img src="https://agentmods.dev/badge/skills/kemetra/seshat-bi/approval-evidence-pack.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 | $0.00051 | $0.04905 |
| Opus 5 | $0.00026 | $0.02452 |
| Sonnet 5 | $0.00010 | $0.00981 |
| Haiku 4.5 | $0.00005 | $0.00490 |
Grade A, and why
approval-evidence-pack 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 11d 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 — 345 lines — stays where its author put it; the contents beside it link to each section on GitHub.
approval-evidence-pack
- Roadmap feature: F035 On-disk spec:
specs/063-approval-evidence-pack/(dir 063 == F035; when the dir number and the F-number disagree, the roadmap F-number wins). - Authority category: Product Module /
artifact-writing(the F024 enumerated declaration -- seedocs/architecture/product-modules.md).
Four of the seven readiness stages -- Mapping Ready, Semantic Model Ready,
Dashboard Ready, Publish Ready -- require a NAMED-HUMAN approval the agent is
constitutionally forbidden to grant for itself (Principle V; recorded in each
table's mappings/<table>/readiness-status.yaml approvals[]). When such a human
is asked to sign a stage gate, the evidence they need is real but SCATTERED: the
per-stage readiness doc says what the gate requires; readiness-status.yaml carries
the four-status state and the open blockers; the AL1 assumption ledger (surfaced
from the metric contracts) may flag an unresolved judgment call; the parked-on map
may show the work is blocked on a shared bottleneck. This skill is the COMPOSER
that assembles those scattered, already-committed artifacts into ONE ordered
PRE-approval pack for one (table, stage). It FILLS the pack; it originates no
evidence and owns no truth. Every line points back at the artifact it came from;
any source that is missing or unfilled is a BLOCKER, never papered over with
invented content. The pack surfaces the gate's readiness picture; it never signs it.
Boundary against neighbouring shipped work (read first)
- F028 evidence-pack-generator (
.claude/skills/evidence-pack-generator/, spec 022) composes a LATE-STAGE, per-table 10-section pack for the Semantic Model -> Dashboard -> Publish window only, written tomappings/<table>/evidence-pack-index.md/evidence-pack-summary.md. This module is a PRE-approval packet GENERIC across ALL seven stage gates (stage parameter), scoped to what ONE gate needs, written tomappings/<table>/approval-evidence-pack-<stage>.md. It REUSES F028's surface-never-assert discipline and empty-approvals rule verbatim in spirit; it does NOT edit F028 and does NOT re-render F028's 10 sections. - F027 Approval Console (
.claude/skills/approval-console/,templates/approval-request.md/approval-decision.md, spec 021) packages ONE raised judgment call into a decidable request and TRANSCRIBES a human's answer back into the committed artifacts. This module packages a WHOLE-GATE readiness picture (many pieces of evidence, not one question) and WRITES NOTHING BACK -- it never transcribes an answer, never appends toapprovals[]. The two compose: this pack is the evidence a human reads BEFORE the Approval Console records their signature.
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.
- 11d ago First seen · 345 lines · 51 tokens per session scan A f443ddb346f2
approval-evidence-pack is a skill published in the GitHub repository Kemetra/Seshat-BI (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 51 tokens to every session and 4,905 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
powerbi-project-management
Project management for BI projects. Use when the user mentions: project plan, sprint, milestone, delivery, roadmap, stakeholder, requirements, scope, backlog, user story, epic, data product, product management, agile, kanban, release plan, go-live, risk, RAID log, data platform delivery, data project, BI delivery…
status-truth
Use when work is leaving the desk — findings, a status, or a number that must hold up in the room. The state of the WORK itself is being reported - a status report or weekly/steering update - and the truth must survive the pull to look green: every claim carries provenance, every green names what would turn it red…
jira_to_bi_spec
Skill "jira_to_bi_spec" from G-Schumacher44/strata-oss, covering skill: jiratobispec, trigger, inputs, allowed tools and forbidden.
ktx-analytics
Use when answering a question that needs data from a ktx-connected database - investigating, analyzing, "how many", "show me", "what's the breakdown of", finding records by value, exploring tables, comparing periods, explaining metrics, or any data-analysis request. Triggers even when the user does not say…
metabase_ingest
Convert Metabase questions, models, and metrics into ktx Semantic Layer source definitions. Covers result-metadata to KSL column type mapping, FK/PK detection, near-duplicate deduplication, pre-aggregation decomposition, join-graph connectivity, and how to react to priorProvenance from earlier ingest syncs. Load when…
sl_capture
How to capture new reusable patterns into ktx's semantic layer - when a measure, segment, or join belongs in the catalog and how to write it generically so it stays small and useful over time. Loaded by the post-turn memory-agent only. The research agent does not write to the SL.