pi-everything (pi extension)

Instant, machine-wide file search for the pi coding agent on Windows, backed by the Everything index.

Everything is a Windows desktop search utility that indexes file and folder names in real time — a filename search across the whole computer returns results in milliseconds, with no directory walk and no waiting on a disk scan. This extension recreates the CodingRookie98/everything-search skill (MIT) as native pi tools wired directly to the index via es.exe.

Tools

| Tool | Purpose | |---|---| | everything_search | Machine-wide instant file search (name/path/extension/size/date/content). Params: query, limit, offset, sort, order, details, count_only, total_size, columns, regex, case_sensitive, whole_word, match_path, diacritics. | | everything_grep | Content search without the directory walk: query selects candidate files from the index (same syntax as everything_search; defaults to the current project, excluding node_modules/.git), then an in-process scan. Params: pattern, query, ignoreCase, literal, context, limit, summary. Output is hashline-anchored (LINE:HASH|content) — byte-compatible with the grep tool, feeds the edit tool. | | everything_folder_size | Recursive folder size from the index (instant, no disk scan). Param: path. | | everything_syntax | Returns the bundled full query-syntax reference (syntax.md) — agents call it when they need syntax beyond basics. Optional include_columns: true also returns the full 55-name property-column allowlist. | | everything_status | Diagnostics: es.exe presence/version/path, service reachability, indexed item count, index freshness (stale-index warning). |

everything_grep output budget. Output is capped at 20 KB and the scan stops as soon as the budget (or limit) is full — candidate files after that are not read, and the marker says so: [everything_grep: output budget (20 KB) reached after N matches in M files. K candidate files not scanned. …]. Match lines over 500 chars render as first 500 + ... [truncated] (parity with the builtin grep tool, plus a footer when any line was cut); the anchor hash is always of the full line, so edits still operate on complete content. Per-file cap: once the running match total passes 50, each file renders at most 10 matches + a ... +K more matches note. summary: true returns per-file counts only (the scan always completes; the budget bounds the file list).

How agents learn the syntax

Three layers, so the agent never has to guess:

1. Always visible — the query parameter description carries a compact syntax cheat sheet (operators, wildcards, size:/dm:/parent:/content:, type macros, examples) in every tool schema.
2. On demand — everything_syntax returns the full reference (bundled syntax.md); pass include_columns: true and it also appends the 55-name property-column list.
3. Self-diagnosis — everything_status + error messages tell the agent exactly what to do when something is off (es missing, service down, stale index).

How agents are steered to prefer it

• Tool description opens with an explicit "PREFER THIS OVER find/ls/glob OR bash directory scans".
• promptSnippet puts a one-liner in the system prompt's Available-tools section.
• promptGuidelines adds a persistent Guidelines bullet to the system prompt of every session (including subagent sessions) that has the tool active: use everything_search first for any file discovery; grep only for contents.

Prerequisites (Windows)

• Everything — installer
• es.exe (Everything CLI) in PATH — CLI download
• Everything service running (start Everything.exe)
• Recommended Everything options (index folder size, USN journal, IPC enabled) — reference screenshots in docs/options-screenshots/
• Tested against the Everything 1.5.0.1423 beta + es.exe 1.1.0.37; 1.4-era es.exe also works (automatic CSV fallback in details mode)

Install

The easiest way — from npm (pi.dev gallery):

pi install npm:pi-everything

(Windows only — Everything is a Windows file index. See Prerequisites below.)

The manual deploy below still works if you prefer clone-and-copy.

Deploy

Copy the extension files into pi's global extensions directory as its own folder:

~/.pi/agent/extensions/everything-search/
├── index.ts
├── grep-core.ts
├── syntax.md
└── README.md

Auto-discovered (~/.pi/agent/extensions/*/index.ts); hot-reloadable with /reload. No package.json/node_modules needed — typebox, @earendil-works/pi-ai, and @earendil-works/pi-coding-agent resolve from pi's own install via jiti.

Implementation notes (verified against es 1.1.0.37 + Everything 1.5.0.1423, 2026-09-10 UTC)

• UTF-8 via export files: es.exe stdout uses the environment OEM codepage (é→0x8A in cp850, CJK→? — lossy). Results are always captured with -export-txt / -export-tsv (verified UTF-8), read as UTF-8, temp file deleted after.
• Sort quirk: direction must be hyphen-joined (-sort size-descending); space-separated (-sort size descending) silently treats the direction as a second search term and returns nothing. The extension always passes an explicit hyphen-joined direction.
• Pagination: -n N fetches N; -viewport-offset only skips within that fetch. The extension fetches offset+limit (capped at 2000) and slices locally.
• Counts: total matches come from -get-result-count (parallel with the results call, both ~50 ms).
• Details: -full-path-and-name -size -date-modified -date-format 1 -no-header -export-tsv gives path<TAB>size-bytes<TAB>ISO-date.
• 1.4 fallback: -export-tsv is unknown to es.exe from the Everything 1.4 era (exit code 6). Details mode then retries with -export-csv (also UTF-8) and converts quoted CSV lines to tab shape, so the extension works with both 1.4-era and 1.5-era es.exe.
• Total size: -get-total-size (parallel, ~50 ms). The 1.5 beta returns 2⁶⁴−1 for zero-result queries; totalSizeOf() maps that sentinel to 0.
• Property columns: -add-columns accepts 55 verified property names (allow-listed in index.ts; es.exe rejects unknown names with Error 4: Unknown column). The full list is in the "Property column names" section below; the curated schema/syntax.md text deliberately omits it so agents don't request useless columns — everything_syntax with include_columns: true returns it on demand. Extra columns imply details mode.
• Broken 1.5 search functions: depth:, child-count:, count:, max-threads:, unique:, distinct:, dupe-min/max/count:, len:, to-number:, hex:, size-on-disk:, si:, and &sp: entities parse but return 0 via es.exe on this build. syntax.md documents them under "Do NOT use with this es.exe build" so agents never try them.
• Multi-term AND contract: space-separated terms are ANDed — but only as SEPARATE es.exe command-line arguments; ONE argument containing spaces is a phrase (literal substring). splitQueryTerms() in index.ts passes query terms as separate arguments (double quotes group a term, e.g. a phrase). No query rewriting is performed.
• Index freshness: everything_status checks the newest file date-modified present in the index (not date-recently-changed, which is unreliable post-rebuild). If nothing has been modified in 24h it warns the index may be stale (USN monitoring off) — on the reference install USN live updates were verified working (2026-09-09) after a forced rebuild (new files indexed within seconds).

Property column names (all 55, verified)

-add-columns accepts exactly these names (es.exe rejects anything else with Error 4: Unknown column). The tool's columns parameter description lists only nine common examples by design; this list is the complete reference. Media/document columns (width, height, length, date-taken, artist, ...) are empty for items that lack the property.

attributes, audio-bitrate, audio-bits-per-sample, audio-channels, audio-sample-rate, aspect-ratio, author, last-author, album, artist, bit-depth, camera-maker, camera-model, comment, company, content-type, date-acquired, date-accessed, date-changed, date-created, date-run, date-recently-changed, date-taken, extension, file-signature, f-stop, frame-rate, genre, hard-link-count, height, iso-speed, kind, latitude, length, line-count, lens-model, longitude, page-count, parent-name, people, product-name, product-version, run-count, slide-count, software, tags, title, track, total-bitrate, video-bitrate, version, volume-label, width, word-count, year

Source attribution

Query syntax reference adapted from CodingRookie98/everything-search (MIT License). Not affiliated with voidtools.