rpce-tool-gotchas SKILL.md skillA

rpce-tool-gotchas from moonray/repoprompt-workflows is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add moonray/repoprompt-workflows/rpce-tool-gotchas@git:20261014.15616ea

Pins git:20261014.15616ea in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
moonray/repoprompt-workflows
Path
.agents/skills/rpce-tool-gotchas/SKILL.md
Versions recorded
3
Latest version
git:20261014.15616ea, committed Oct 14, 2026
First version recorded
Oct 10, 2026
Size
4,427 bytes
Repository stars
11
License
MIT

RepoPrompt CE workspace-tool boundaries that agents keep re-learning by failure. Use when an RPCE file/edit tool (apply_edits, file_actions, read_file, MCP file writes) fails or cannot see a path outside the loaded workspace roots — /tmp one-off scripts, a sibling repo, ~/ dotfiles: those tools are root-scoped by design, so route out-of-root reads/writes through Bash (heredoc/python) from the start. Also use when stacking more edits onto files just changed by a batch of edits, especially…

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# RPCE tool gotchas

Three verified facts about RepoPrompt CE's workspace tools, each costing real rework before it was encoded here (provenance: cross-workspace session mining, 2026-10 for §1–§2; single-session mining 2026-10-11 with three incidents for §3 — general facts, no single repo).

## 1. Workspace file tools are root-scoped

`apply_edits`, `file_actions`, and the MCP read/write tools operate ONLY on paths inside the loaded workspace root(s). A path outside them — `/tmp` throwaway scripts, a sibling repository, `~/.config` — either fails outright ("file does not exist") or, worse, strands work when a create is assumed to have happened but silently didn't.

**Do:** when a task needs files outside loaded roots (one-off evaluator scripts, cross-repo ledger edits, scratch dumps), write them via Bash from the start — heredoc into a file, or `python3 - <<'EOF'` for anchored edits — and verify by reading back. Don't first try the workspace tool and discover the boundary by failure: the failure reads as "file does not exist" and invites recreating content that was never lost, or losing content that was never written.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add moonray/repoprompt-workflows/rpce-tool-gotchas@sha256:c2292fd1713cc671

The line at the top of this page pins the label git:20261014.15616ea, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_zgy5oh2ftyuve6d4.svg)](https://markdownregistry.com/a/art_zgy5oh2ftyuve6d4)

3 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 3 versions of this file, between Oct 10, 2026 and Oct 14, 2026, about one every 2 days.

versioncommittedcommitsizeaudit
git:20261014.15616ea latest2026-10-14 15616ea 4,427 BA view · diff
git:20261011.8064e642026-10-11 8064e64 4,083 BA view · diff
git:20261010.ccd3fa72026-10-10 ccd3fa7 2,914 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (4427 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

moonray/repoprompt-workflows · 11 stars · license MIT · pushed 2026-10-16 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_zgy5oh2ftyuve6d4
GET https://markdownregistry.com/api/v1/resolve?ref=moonray/repoprompt-workflows/rpce-tool-gotchas
GET https://markdownregistry.com/api/v1/blob/c2292fd1713cc671cfb93ff98034b32245ebf9d55ad2d61aa19126f9cbf92065

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from moonray/repoprompt-workflows

bot-wall SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/bot-wall/SKILL.md · Use whenever an HTTP fetch of an external site is blocked by a bot wall — 403/Access Denied/"Request unsuccessful" on a…
git:20261010.4e23de3 · audit A · 11 stars
chat-mining SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/chat-mining/SKILL.md · Use when asked to mine or analyze past agent session chats/transcripts (e.g. "what can we learn from workspace X's…
git:20261010.9ca2554 · audit A · 11 stars
chrome SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/chrome/SKILL.md · Use whenever driving or visiting a website with browser automation — the chrome-devtools MCP (dev chrometools…
git:20261010.4e23de3 · audit A · 11 stars
document SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/document/SKILL.md · Use when reconciling documentation to code changes or auditing documentation drift in a repo. Applies whenever the user…
git:20260822.91c6099 · audit A · 11 stars
live-dryrun SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/live-dryrun/SKILL.md · Use when the user asks for "automated user testing", a "dry run", or "live verification" of a backend/API/provider…
git:20261010.d8782ee · audit A · 11 stars
maintainability-review SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/maintainability-review/SKILL.md · Use for a strict, quality-only structural review that hunts for dramatic simplifications, not local cleanup. Trigger on…
git:20260710.161b2ae · audit A · 11 stars
model-routing SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/model-routing/SKILL.md · Use when choosing an oracle model or effort level for spec, plan, or review consultations (ask_oracle, context_builder…
git:20261011.a5e82c5 · audit A · 11 stars
review-depth SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/review-depth/SKILL.md · Use when deciding how much review effort a change warrants — quick, standard, or deep — picked from the change's size…
git:20260710.161b2ae · audit A · 11 stars
review-quality SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/review-quality/SKILL.md · Use when producing, consuming, triaging, or revalidating code-review findings — review artifacts, P0/P1 findings, fix…
git:20261007.59c4e76 · audit A · 11 stars
spec-conformance SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/spec-conformance/SKILL.md · Use when closing a spec-driven feature/issue or auditing whether an implementation actually matches its spec. Given a…
git:20260710.161b2ae · audit A · 11 stars
spec-plan-readiness SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/spec-plan-readiness/SKILL.md · Use before implementation when a Spec and Deep Plan must be checked for coding readiness. Applies a deterministic…
git:20260710.161b2ae · audit A · 11 stars
spec-quality SKILL.md skill
moonray/repoprompt-workflows · .agents/skills/spec-quality/SKILL.md · Use when drafting, reviewing, repairing, or validating scenario-based specs, product specs, implementation specs…
git:20260710.161b2ae · audit A · 11 stars

Every file in moonray/repoprompt-workflows

Browse by kind, by grade A, or by owner.