How do I install the Find skill?

Run npx skillmds@latest add vslipchenko/find in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Find?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

vslipchenko (@vslipchenko) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Find by vslipchenko — Agent Skill

Name: Find
Author: vslipchenko

Find

Look up one or more records in the estimator history by Jira key. Use when the user wants to find / show / inspect a stored ticket record (e.g. "find ABC-7777 in history", "show the record for 7777", "look up ABC-1, ABC-2").

vslipchenko d469b04 5 files · 11.9 KB Updated Sep 17, 2026

File contents

Estimator — Find History Records

Look up records in history.csv by Jira key and show their full stored fields.

Data location

Resolve <DATA_DIR> as the other skills do — the per-user data dir ~/.estimator:

Bash: DATA_DIR="$HOME/.estimator"
PowerShell: $DATA_DIR = Join-Path $env:USERPROFILE '.estimator'

Step 1 — Preflight

Resolve <DATA_DIR>. If <DATA_DIR>/history.csv does not exist, tell the user to run /sync first to build the dataset, and stop.

Step 2 — Resolve the key(s)

Take one or more keys from the user's message/argument (space- or comma-separated) and uppercase them. Expand bare numbers via config.json project_keys: exactly one cached key → prefix it (7777 → ABC-7777); several → ask which project; none cached → ask for the project key. If no key can be determined, prompt the user for the ticket key(s).

Step 3 — Query the history

Run the lookup with the first available runtime:

python3 "<PLUGIN_ROOT>/skills/find/scripts/find_records.py" "<DATA_DIR>/history.csv" KEY [KEY ...]
else python "<PLUGIN_ROOT>/skills/find/scripts/find_records.py" "<DATA_DIR>/history.csv" KEY [KEY ...]
else py -3 "<PLUGIN_ROOT>/skills/find/scripts/find_records.py" "<DATA_DIR>/history.csv" KEY [KEY ...]
else node "<PLUGIN_ROOT>/skills/find/scripts/find_records.mjs" "<DATA_DIR>/history.csv" KEY [KEY ...]
else fallback: read <DATA_DIR>/history.csv yourself (treat it as ;-delimited with RFC 4180 quoting — a record may span multiple physical lines inside a quoted cell) and select records whose first column equals a requested key, tracking each match's 1-based data-row position (header excluded).

The script prints JSON: {"found": [{"row": N, ...all fields}], "missing": [...]}.

Resolving <PLUGIN_ROOT>: $CLAUDE_PLUGIN_ROOT is usually empty inside a skill's shell, so don't rely on it. <PLUGIN_ROOT> is the directory two levels above this skill's announced base directory (the base dir shown when this skill loaded is …/skills/<name>, so its parent's parent is the plugin root). Substitute that absolute path for <PLUGIN_ROOT> in the commands above.

Step 4 — Present

For each found record show its row number and all fields as Label: value, using the human-readable column names (Key, Parent, Issue Type, Title, Description, Components, Labels, Story Points, Time Spent [s], Resolution Date, Comments, Design Link, Estimate Basis). Then list any missing keys plainly. If nothing was found, say so.

Error handling

No history.csv → guide the user to run /sync; do not fabricate data.
No / unparseable keys → prompt.
Script runtime unavailable → use the Step-3 fallback.

---name: finddescription: Look up one or more records in the estimator history by Jira key. Use when the user wants to find / show / inspect a stored ticket record (e.g. "find ABC-7777 in history", "show the record for 7777", "look up ABC-1, ABC-2").---# Estimator — Find History RecordsLook up records in `history.csv` by Jira key and show their full stored fields.## Data locationResolve `<DATA_DIR>` as the other skills do — the per-user data dir `~/.estimator`:- Bash: `DATA_DIR="$HOME/.estimator"`- PowerShell: `$DATA_DIR = Join-Path $env:USERPROFILE '.estimator'`## Step 1 — PreflightResolve `<DATA_DIR>`. If `<DATA_DIR>/history.csv` does not exist, tell the userto run `/sync` first to build the dataset, and stop.## Step 2 — Resolve the key(s)Take one or more keys from the user's message/argument (space- orcomma-separated) and uppercase them. Expand bare numbers via `config.json``project_keys`: exactly one cached key → prefix it (`7777` → `ABC-7777`);several → ask which project; none cached → ask for the project key. If no keycan be determined, prompt the user for the ticket key(s).## Step 3 — Query the historyRun the lookup with the first available runtime:1. `python3 "<PLUGIN_ROOT>/skills/find/scripts/find_records.py" "<DATA_DIR>/history.csv" KEY [KEY ...]`2. else `python "<PLUGIN_ROOT>/skills/find/scripts/find_records.py" "<DATA_DIR>/history.csv" KEY [KEY ...]`3. else `py -3 "<PLUGIN_ROOT>/skills/find/scripts/find_records.py" "<DATA_DIR>/history.csv" KEY [KEY ...]`4. else `node "<PLUGIN_ROOT>/skills/find/scripts/find_records.mjs" "<DATA_DIR>/history.csv" KEY [KEY ...]`5. else **fallback**: read `<DATA_DIR>/history.csv` yourself (treat it as `;`-delimited with RFC 4180 quoting — a record may span multiple physical lines inside a quoted cell) and select records whose first column equals a requested key, tracking each match's 1-based data-row position (header excluded).The script prints JSON:`{"found": [{"row": N, ...all fields}], "missing": [...]}`.**Resolving `<PLUGIN_ROOT>`:** `$CLAUDE_PLUGIN_ROOT` is usually empty inside askill's shell, so don't rely on it. `<PLUGIN_ROOT>` is the directory **two levelsabove this skill's announced base directory** (the base dir shown when this skillloaded is `…/skills/<name>`, so its parent's parent is the plugin root).Substitute that absolute path for `<PLUGIN_ROOT>` in the commands above.## Step 4 — PresentFor each found record show its **row number** and **all fields** as`Label: value`, using the human-readable column names (Key, Parent, Issue Type,Title, Description, Components, Labels, Story Points, Time Spent [s], ResolutionDate, Comments, Design Link, Estimate Basis). Then list any **missing** keysplainly. If nothing was found, say so.## Error handling- No `history.csv` → guide the user to run `/sync`; do not fabricate data.- No / unparseable keys → prompt.- Script runtime unavailable → use the Step-3 fallback.

Frequently asked questions