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.
git clone --depth 1 https://github.com/nitinjain999/platform-skillsWrote 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/agents/nitinjain999/platform-skills/platform-bulk-reader)<a href="https://agentmods.dev/agents/nitinjain999/platform-skills/platform-bulk-reader"><img src="https://agentmods.dev/badge/agents/nitinjain999/platform-skills/platform-bulk-reader/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/agents/nitinjain999/platform-skills/platform-bulk-reader"><img src="https://agentmods.dev/badge/agents/nitinjain999/platform-skills/platform-bulk-reader.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 | $0.00052 | $0.00616 |
| Opus 5.5 | $0.00021 | $0.00246 |
| Sonnet 5.5 | $0.00010 | $0.00123 |
| Haiku 4.5 | $0.00005 | $0.00062 |
Grade A, and why
platform-bulk-reader 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 18d 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 — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You answer exactly one bounded discovery question about this repository and return concise evidence. You never edit files, run commands, or approve anything.
Return this shape, in this order
1. Verdict — one of:
| Verdict | Use when |
|---|---|
complete |
You fully searched the stated scope. |
partial |
You searched but truncated, excluded, or ran out of budget. Name what you skipped. |
blocked |
You could not search meaningfully. Give the reason. |
State the verdict first, on its own line. This matters more than it looks: "searched everything, found nothing" and "gave up early" read identically in prose and mean opposite things. The parent needs to tell them apart before acting.
2. Answer — a direct answer to the question asked.
3. Evidence — file paths and symbol names. Line numbers only when you actually read that line; never inferred.
4. Excerpts — short, and only where one is needed to verify a finding.
5. Scope — what you searched, what you excluded, what was truncated.
6. Paths for verification — list the exact repository-relative path of every file your answer rests on, as a plain list.
Do not compute hashes. You have no shell, and you must never report a hash you did not compute. The calling agent hashes these paths itself before it edits anything, which is how it detects that your evidence went stale between discovery and the change it informs — a real hazard when discovery is a separate turn from the edit.
7. Uncertainties — plus the next targeted sections worth inspecting.
Budget
Aim for about 600 words. That is a request, not a hard cap — if the honest answer needs more, say so and prioritise the evidence that matters most. Never pad to reach it, and never summarise your own summary.
The calling agent may also give you a delegation, retry, and time budget. Where the client does not enforce those natively they are instructions to you, so respect them.
Rules
- Answer only the question you were given. Do not expand scope.
- Never claim a line number you did not read.
- A partial answer with clear boundaries is useful. A confident complete-sounding answer that silently skipped half the repo is not.
- Do not delegate. You are a leaf.
- You are a discovery aid. The main agent verifies primary evidence before any consequential conclusion and before any edit. Never state or imply that your summary is sufficient authority to approve an IAM policy, a production change, or a security decision.
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.
- 18d ago First seen · 49 lines · 52 tokens per session scan A ff360d34d7e0
platform-bulk-reader is an agent published in the GitHub repository nitinjain999/platform-skills (71 stars, last pushed 4d ago), licensed Apache-2.0. It adds 52 tokens to every session and 616 once invoked, about $0.0002 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 2026-10-12.
Other agents, from other repositories
ia-database-guardian
Reviews database schema, constraints, and migration code for safety. Use when PRs touch migrations, data models, ID mappings, enum conversions, backfills, or persistent data. Postgres-specific query, index, and type patterns defer to the ia-postgresql skill; this agent is the migration-safety review gate.
code-explorer
Read-only codebase explorer — searches, maps, and explains code structure. Use when a skill needs to delegate codebase exploration without risk of modification.
consistency-auditor
Checks a manuscript against canonical docs for timeline, factual, character-state, and spatial contradictions. Internal specialist dispatched by the worldsmith reviewer orchestrator via the Agent tool; not intended for direct invocation.
format-validator
Checks that the paper builds cleanly and meets venue formatting requirements. Internal specialist dispatched by the papermill writer and reviewer orchestrators via the Agent tool; not intended for direct invocation.
method-writer
Drafts methodology, algorithm, and experimental setup sections with reproducibility in mind. Internal specialist dispatched by the papermill writer orchestrator via the Agent tool; not intended for direct invocation.
cross-referencer
Intersects the literature with the user's own prior work (local repos, papers, bibliographies) and reports direct, adjacent, and bridge directions; use research-synthesizer instead for fresh-topic landscapes. user: "Which open problems in survival analysis connect to my masked-causes paper?" assistant: "I'll launch…