Skip to content
Agentmart
Submit a tool

ingest

Bootstraps the memory store from existing material (repos, PDFs, transcripts).

Skill by fjwood69

IndexedAGPL-3.0★ 32
Write a review

Chalk Labs

Need help deploying ingest?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/fjwood69-ingest

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Bootstraps the memory store from existing material (repos, PDFs, transcripts). Use to seed memory from a project or document.

Use cases: documents

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/ingest

What's inside

Skill name
ingest
Description
Bootstraps the memory store from existing material (repos, PDFs, transcripts). Use to seed memory from a project or document.
License
AGPL-3.0
Instructions
103 lines in SKILL.md · view on GitHub

Before installing

  • Instructions call the network

Files (1)

  • SKILL.md

SKILL.md (first lines)

## Filesystem mode (server-local paths)

1. If the user passes `--status`, call `mori-ingest_status` and present the table.
2. If the user passes `--preview` or `--dry-run` with no source provided, remind them to provide `--source <path>`.
3. For `--preview`: call `mori-ingest_preview` with the `--source` path(s) and any `--type` or `--since` arguments. Present the chunk breakdown and cost estimate. Remind the user this is zero-cost — no LLM was called.
4. For `--dry-run`: call `mori-ingest` with `dry_run=true` and all provided arguments. Report what would be written and the actual cost incurred (the LLM was called, just nothing committed).
5. For a real ingestion run: call `mori-ingest` with all provided arguments. Report: sources processed, chunks sent, memories written, cost estimate, errors.
6. If errors occurred for specific files, report them individually.
7. After successful ingestion, suggest running `/brief` to reload shared memories so the new entries are visible.
8. For large or expensive-looking sources, suggest `/ingest --preview` first to check before committing.

## Content mode (remote clients — HTTP upload)

When the user passes `--content <path>`, upload the file directly to the mori-ingestion pod.
Do NOT use the `mori-ingest_content` MCP tool — it has been removed. Use HTTP multipart upload only.

Requires:
- `MORI_INGEST_URL` env var (set by installer — NOT derived from `MORI_URL`)
- `MORI_API_KEY` env var (same key as MCP auth)

### Linux/macOS

```bash
JOB=$(curl -sf -X POST "${MORI_INGEST_URL}/api/ingest/upload" \
  -H "X-Api-Key: ${MORI_API_KEY}" \
  -F "files=@${path}" \
  -F "focus=${focus}" \
  -F "tier=${tier}" \
  -F "tags=${tags}" \
  -F "dry_run=${dry_run}")
JOB_ID=$(echo "$JOB" | python3 -c "import sys,json; print(json.load(sys.stdin)['job_id'])")
```

### Windows (PowerShell)

```powershell
$r = Invoke-RestMethod -Uri "$env:MORI_INGEST_URL/api/ingest/upload" `
  -Method POST `
  -Headers @{"X-Api-Key" = $env:MORI_API_KEY} `
  -Form @{
    files   = Get-Item $path
    focus   = $focus
    tier    = $tier
    tags    = $tags
    dry_run = $dry_run
  }
$JOB_ID = $r.job_id
```

### Polling (both platforms)

Poll every 5 seconds until `status` is `complete` or `failed`:

```bash
# Linux/macOS
while true; do
  STATUS=$(curl -sf "${MORI_INGEST_URL}/api/ingest/job/${JOB_ID}" \
    -H "X-Api-Key: ${MORI_API_KEY}")
  STATE=$(echo "$STATUS" | python3 -c "import sys,json; print(json.load(sys.stdin)['status'])")
  [ "$STATE" = "complete" ] || [ "$STATE" = "failed" ] && break
  sleep 5
done
echo "$STATUS"
```

```powershell
# Windows
do {
  $status = Invoke-RestMethod -Uri "$env:MORI_INGEST_URL/api/ingest/job/$JOB_ID" `
    -Headers @{"X-Api-Key" = $env:MORI_API_KEY}
  if ($status.status -notin "queued","running") { break }
  Start-Sleep 5
} while ($true)
$status
```

Report: memories written, estimated cost, errors. After success, suggest `/brief` to reload shared memories.

### Error handling

- `MORI_INGEST_URL` not set → tell user to set it (e.g. `http://localhost:8969`) and re-run
- HTTP 413 → file too large (50MB/file, 200MB total)
- HTTP 401 → API key mismatch — check `MORI_API_KEY`
- status `failed` → report `errors` field from job response

## Argument mapping

| User flag | Action |
|-----------|--------|
| `--source <path>` (repeatable) | `mori-ingest` MCP tool → `source` list |
| `--content <path>` | HTTP multipart upload to `MORI_INGEST_URL` |
| `--type <type>` | `type` (auto, transcripts, git, docs, image) |
| `--focus <area>` | `focus` (all, decisions, architecture, conventions, gotchas) |
| `--tier <tier>` | `tier` (working, canonical, ephemeral) |
| `--tags <tags>` | `tags` (comma-separated string) |
| `--since <duration>` | `since` (e.g. "30d", "90d") — filesystem mode only |
| `--dry-run` | `dry_run=true` |
| `--force` | `force=true` |
| `--max-cost <amount>` | `max_cost` (float, USD) — filesystem mode only |
| `--preview` | call `mori-ingest_preview` instead of `mori-ingest` |
| `--status` | call `mori-ingest_status` |

From fjwood69/mori, AGPL-3.0.

Signals

Stars
32
Forks
5
Open issues
14
Last commit (this folder)
Jun 5, 2026 (4 months ago)
Last release
Oct 3, 2026 (28 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
AGPL-3.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

Same skill elsewhere

Copies of this SKILL.md published by others; search shows only one of them.

More from fjwood69

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get ingest running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying ingest?

CallGet help
Cite this page

ingest skill. Agent Mart. https://agentmart.dev/skills/fjwood69-ingest (last updated 2026-10-24)