jaz-clio 5.30.1 → 5.30.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +66 -48
- package/assets/skills/api/SKILL.md +1 -1
- package/assets/skills/api/references/endpoints.md +11 -1
- package/assets/skills/cli/SKILL.md +2 -2
- package/assets/skills/conversion/SKILL.md +1 -1
- package/assets/skills/jaz-kit/SKILL.md +1 -1
- package/assets/skills/jaz-pseudo-sql/SKILL.md +1 -1
- package/assets/skills/jobs/SKILL.md +1 -1
- package/assets/skills/transaction-recipes/SKILL.md +1 -1
- package/cli.mjs +408 -401
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -1,4 +1,4 @@
|
|
|
1
|
-
# Clio
|
|
1
|
+
# Clio
|
|
2
2
|
|
|
3
3
|
<p align="center">
|
|
4
4
|
<a href="https://www.npmjs.com/package/jaz-clio"><img src="https://img.shields.io/npm/v/jaz-clio?style=for-the-badge&logo=npm" alt="npm"></a>
|
|
@@ -7,63 +7,65 @@
|
|
|
7
7
|
<a href="https://github.com/teamtinvio/jaz-ai/blob/main/LICENSE"><img src="https://img.shields.io/github/license/teamtinvio/jaz-ai?style=for-the-badge&color=green" alt="License"></a>
|
|
8
8
|
</p>
|
|
9
9
|
|
|
10
|
-
|
|
10
|
+
<p align="center"><b>Jaz accounting on the command line, and inside any AI agent.</b></p>
|
|
11
|
+
|
|
12
|
+
357 tools · 66 command groups · 7 skills · 13 calculators · 12 close playbooks · 158 field-tested API rules.
|
|
11
13
|
|
|
12
14
|
```bash
|
|
13
15
|
npm install -g jaz-clio
|
|
14
16
|
```
|
|
15
17
|
|
|
16
|
-
|
|
18
|
+
Node.js 18+. Works with [Jaz](https://jaz.ai) and [Juan Accounting](https://juan.ac) on the same API.
|
|
17
19
|
|
|
18
20
|
## Contents
|
|
19
21
|
|
|
20
|
-
- [Three
|
|
21
|
-
- [CLI](#cli)
|
|
22
|
-
- [MCP
|
|
23
|
-
- [Skills](#skills)
|
|
24
|
-
- [
|
|
25
|
-
- [
|
|
22
|
+
- [Three ways in](#three-ways-in)
|
|
23
|
+
- [CLI](#cli)
|
|
24
|
+
- [MCP server](#mcp-server)
|
|
25
|
+
- [Skills](#skills)
|
|
26
|
+
- [Jaz Kit · run your practice](#jaz-kit--run-your-practice)
|
|
27
|
+
- [Auth](#auth)
|
|
28
|
+
- [Semantic help-center search](#semantic-help-center-search-optional)
|
|
29
|
+
- [Privacy](#privacy) · [Support](#support) · [License](#license)
|
|
26
30
|
|
|
27
|
-
## Three
|
|
31
|
+
## Three ways in
|
|
28
32
|
|
|
29
|
-
| | What
|
|
33
|
+
| | What it is | Try it |
|
|
30
34
|
|---|---|---|
|
|
31
35
|
| **CLI** | Every accounting operation as a command | `clio invoices list` |
|
|
32
|
-
| **MCP** |
|
|
33
|
-
| **Skills** | Teach any agent the Jaz API | `clio init` |
|
|
36
|
+
| **MCP** | A local server for Claude Code, Cursor, Codex, Copilot | `clio mcp` |
|
|
37
|
+
| **Skills** | Teach any agent the Jaz API, no server needed | `clio init` |
|
|
34
38
|
|
|
35
|
-
|
|
39
|
+
On top of these, one layer for accountants closing real books across many companies: **[Jaz Kit](#jaz-kit--run-your-practice)**.
|
|
36
40
|
|
|
37
41
|
## CLI
|
|
38
42
|
|
|
39
43
|
```bash
|
|
40
|
-
clio invoices create --contact "ACME" --json
|
|
41
|
-
clio bank import statement.csv
|
|
42
|
-
clio reports pdf profit-loss
|
|
43
|
-
clio calc lease --payment 5000 --term 36 --rate 5
|
|
44
|
-
clio jobs month-end --period 2026-03
|
|
45
|
-
clio magic create --file receipt.pdf
|
|
46
|
-
clio
|
|
47
|
-
clio invoices search --query 'status:unpaid AND $500+' # Structured per-entity search
|
|
48
|
-
clio invoices search --query 'status:unpaid' --view lean # Compact summary rows (id + key fields), then drill in with get
|
|
44
|
+
clio invoices create --contact "ACME" --json # draft an invoice, JSON back
|
|
45
|
+
clio bank import statement.csv # import and auto-reconcile
|
|
46
|
+
clio reports pdf profit-loss # download the P&L as a PDF
|
|
47
|
+
clio calc lease --payment 5000 --term 36 --rate 5 # IFRS 16, offline, instant
|
|
48
|
+
clio jobs month-end --period 2026-03 # step-by-step close playbook
|
|
49
|
+
clio magic create --file receipt.pdf # AI extracts, drafts the transaction
|
|
50
|
+
clio invoices search --query 'status:unpaid AND $500+' # structured per-entity search
|
|
49
51
|
```
|
|
50
52
|
|
|
51
|
-
66 command groups
|
|
53
|
+
66 command groups, 16 report types, 13 calculators, 12 job playbooks. Every command takes `--json`. Run `clio --help` for the full list.
|
|
52
54
|
|
|
53
|
-
|
|
55
|
+
## MCP server
|
|
54
56
|
|
|
55
|
-
|
|
57
|
+
357 tools for any AI agent that speaks MCP. Runs locally: no cloud, no ports.
|
|
56
58
|
|
|
57
|
-
|
|
59
|
+
> **No install at all?** Claude.ai, ChatGPT, Cowork, and Microsoft Copilot Studio can use Jaz through the hosted connector. Add `https://mcp.jaz.ai/mcp` as a custom connector and sign in with OAuth, no key. The local setup below is for terminal use, scripting, and editors that run MCP servers as local processes.
|
|
58
60
|
|
|
59
|
-
|
|
61
|
+
**Claude Code**
|
|
60
62
|
|
|
61
|
-
**Claude Code:**
|
|
62
63
|
```bash
|
|
63
64
|
claude mcp add jaz -- npx jaz-clio mcp
|
|
64
65
|
```
|
|
65
66
|
|
|
66
|
-
**Cursor
|
|
67
|
+
**Cursor · VS Code · Windsurf**
|
|
68
|
+
|
|
67
69
|
```json
|
|
68
70
|
{
|
|
69
71
|
"mcpServers": {
|
|
@@ -76,45 +78,61 @@ claude mcp add jaz -- npx jaz-clio mcp
|
|
|
76
78
|
}
|
|
77
79
|
```
|
|
78
80
|
|
|
79
|
-
|
|
81
|
+
Several companies at once: comma-separate the keys, or use a personal access token.
|
|
82
|
+
|
|
80
83
|
```json
|
|
81
84
|
{ "env": { "JAZ_API_KEY": "jk-org1-key,jk-org2-key" } }
|
|
82
85
|
```
|
|
83
86
|
|
|
84
|
-
---
|
|
85
|
-
|
|
86
87
|
## Skills
|
|
87
88
|
|
|
88
|
-
158 API rules from production testing: field-name maps, error-recovery patterns, response-shape quirks
|
|
89
|
+
158 API rules from production testing: field-name maps, error-recovery patterns, response-shape quirks, plus 12 job playbooks. Installable into any agent project, no server involved.
|
|
89
90
|
|
|
90
91
|
```bash
|
|
91
|
-
clio init
|
|
92
|
-
clio init --platform cursor
|
|
93
|
-
clio init --no-rules
|
|
92
|
+
clio init # auto-detect the agent, install skills + agent-rules
|
|
93
|
+
clio init --platform cursor # explicit platform
|
|
94
|
+
clio init --no-rules # skills only, skip the agent-rules file
|
|
94
95
|
```
|
|
95
96
|
|
|
96
|
-
|
|
97
|
+
`init` detects the agent (Claude Code, Codex, Copilot, Cursor, Antigravity, Gemini, Windsurf, Goose) and installs the right skill files. It also writes a one-page `jaz-agent-rules.md` to the file your platform reads on open (`CLAUDE.md`, `AGENTS.md`, `.github/copilot-instructions.md`, `.cursor/rules/jaz.mdc`, `.windsurf/rules/jaz.md`, or `GEMINI.md`), so every session starts with the tool-discovery flow and the API gotchas already loaded.
|
|
97
98
|
|
|
98
|
-
|
|
99
|
+
## Jaz Kit · run your practice
|
|
99
100
|
|
|
100
|
-
|
|
101
|
+
The workspace layer for closing real books, whether that is one company or fifty.
|
|
101
102
|
|
|
102
|
-
|
|
103
|
-
|
|
104
|
-
|
|
103
|
+
A close is not one conversation. Month-end runs many steps over several days, and an accountant serving eight clients runs it eight times, with eight different sets of bank accounts, materiality thresholds, and recurring entries. Jaz Kit gives each company a folder that remembers all of it, so no session re-asks what it should already know.
|
|
104
|
+
|
|
105
|
+
```
|
|
106
|
+
/jk-setup create a company workspace, connect its key
|
|
107
|
+
/jk-open acme load its context, verify the connection
|
|
108
|
+
/jk-close 2026-06 run the close, resumable across sessions
|
|
109
|
+
/jk-review approve the drafts waiting on you
|
|
110
|
+
/jk-status every company: what is due, what is pending
|
|
111
|
+
/jk-exit wrap up and journal the session
|
|
105
112
|
```
|
|
106
113
|
|
|
107
|
-
|
|
114
|
+
Also `/jk-keys`, `/jk-policy`, `/jk-teach`, `/jk-save`, `/jk-help`. **`/jaz-*` runs a single workflow; `/jk-*` runs your practice.**
|
|
115
|
+
|
|
116
|
+
Each company lives under `~/Documents/Jaz Kit/orgs/<company>/`, holding its close config, its policies, and its own API key in a gitignored `.env`. A Jaz key is scoped to one company, so the folder you open is the company you work on: nothing to switch, and no way to post to the wrong books once a folder's key checks out.
|
|
117
|
+
|
|
118
|
+
Everything is drafted first, every record carries a link into Jaz for you to review, and an interrupted close resumes exactly where it stopped. Multi-company work needs this CLI, which you already have. Full guide in the [repository README](https://github.com/teamtinvio/jaz-ai#jaz-kit--run-your-practice).
|
|
119
|
+
|
|
120
|
+
## Auth
|
|
121
|
+
|
|
122
|
+
```bash
|
|
123
|
+
clio auth add <api-key> # from Settings → API keys in Jaz
|
|
124
|
+
clio auth whoami # verify
|
|
125
|
+
```
|
|
108
126
|
|
|
109
|
-
|
|
127
|
+
Or set `JAZ_API_KEY` in your environment for scripts and CI. For several companies from the CLI, register each with `clio auth add` and pass `--org <label>` per command, or let Jaz Kit keep one key per company folder for you. Every command takes `--json` for structured output.
|
|
110
128
|
|
|
111
|
-
##
|
|
129
|
+
## Semantic help-center search (optional)
|
|
112
130
|
|
|
113
|
-
`search_help_center`
|
|
131
|
+
`search_help_center` runs keyword search over the bundled help-center corpus by default. Set `CLIO_HELP_CENTER_OPENAI_API_KEY` to add semantic search, which matches on intent rather than exact keywords: the CLI embeds only your query through the OpenAI embeddings API (`text-embedding-3-small`, the model the bundled index was built with) and merges both rankings. On an auth failure it warns once and falls back to keyword search. Use a project-scoped key restricted to embedding models with a low monthly cap. CLI only: MCPB installs ship without the embedding index and always use keyword search.
|
|
114
132
|
|
|
115
133
|
## Privacy
|
|
116
134
|
|
|
117
|
-
Runs on your machine. Calls go to the Jaz API over HTTPS. No telemetry
|
|
135
|
+
Runs on your machine. Calls go to the Jaz API over HTTPS. No telemetry, no data collection.
|
|
118
136
|
|
|
119
137
|
## Support
|
|
120
138
|
|
|
@@ -2380,7 +2380,17 @@ Batch-record judgment entries (1-100 per call). Per-entry independent: acks come
|
|
|
2380
2380
|
{ "data": { "records": [{ "resourceId": "...", "replayed": false, "duplicateCount": 0 }] } }
|
|
2381
2381
|
```
|
|
2382
2382
|
|
|
2383
|
-
`kind`: CLASSIFICATION, MATCH, SCOPE, ASSUMPTION, RISK, METHOD, RECOVERY, DEVIATION, NOTE (the neutral fallback for a judgment logged without a declared type; a missing or blank `kind` defaults to NOTE and is flagged, not rejected). `tier`: LOW, MEDIUM, HIGH, CRITICAL. `refs` entries are OBJECTS: the string grammar `TYPE:resourceId[#field][:RELATION]` travels in `raw` (an unparseable ref is stored with `parsed: false`, never bounced). Optional fields: `ruledOut`, `frame`, `confidence`, `citedRule`, `workflowLabel`, `agentLabel`. `idempotencyKey` makes retries safe: a replay returns the existing entry with `replayed: true`.
|
|
2383
|
+
`kind`: CLASSIFICATION, MATCH, SCOPE, ASSUMPTION, RISK, METHOD, RECOVERY, DEVIATION, NOTE (the neutral fallback for a judgment logged without a declared type; a missing or blank `kind` defaults to NOTE and is flagged, not rejected — never declare NOTE deliberately). `tier`: LOW, MEDIUM, HIGH, CRITICAL. `refs` entries are OBJECTS: the string grammar `TYPE:resourceId[#field][:RELATION]` travels in `raw` (an unparseable ref is stored with `parsed: false`, never bounced). Optional fields: `ruledOut`, `frame`, `confidence`, `citedRule`, `workflowLabel`, `agentLabel`. `idempotencyKey` makes retries safe: a replay returns the existing entry with `replayed: true`.
|
|
2384
|
+
|
|
2385
|
+
**Jot doctrine (fill fields consistently — the server re-scores tier from kind + refs, and declared-vs-computed agreement is a review signal):**
|
|
2386
|
+
|
|
2387
|
+
- **When**: log a judgment when you chose among real alternatives and a write followed, or when you deliberately decided NOT to write. Skip mechanical actions. Jot AFTER the write succeeds; carry the written record's resourceId in `refs`.
|
|
2388
|
+
- **Tier anchors** (mirror the server's rules): CRITICAL = money leaves (`PAY` ref), data destroyed (`DELETE` ref), external send or period lock (`FINALIZE` ref), or a RECOVERY that still drove a write. HIGH = the withheld write (RECOVERY with no mutation ref — it pins via withheld-write, not tier), or RISK/MATCH backed by a write. LOW = a DEVIATION detached from any write. MEDIUM = everything else.
|
|
2389
|
+
- **Kind boundaries**: where a value LANDS (account, tax code) = CLASSIFICATION; how it is COMPUTED = METHOD. Filling one missing fact = ASSUMPTION; drawing a set boundary = SCOPE (carry `frame`). A decided omission after failure = RECOVERY, never DEVIATION.
|
|
2390
|
+
- **Refs relation** is load-bearing: state what the write did (CREATE/UPDATE/DELETE/FINALIZE/PAY/RECONCILE/TRIGGER); SUBJECT only for no-write entries. PAY/DELETE/FINALIZE pin the jot regardless of declared tier.
|
|
2391
|
+
- **confidence**: HIGH = clear rule or precedent; MEDIUM = pattern inference; LOW = a guess a reviewer should check.
|
|
2392
|
+
- **workflowLabel**: use a canonical job name when one fits (`month-end-close`, `quarter-end-close`, `year-end-close`, `bank-recon`, `gst-vat-filing`, `payment-run`, `credit-control`, `supplier-recon`, `audit-prep`, `fa-review`, `document-collection`, `statutory-filing`), else short kebab-case.
|
|
2393
|
+
- **Style**: tight, factual, plain punctuation; one line per field; never repeat content across fields; a call without a `why` is half a record. `duplicateCount > 0` on the ack = already recorded — do not re-jot; search first on repeated workflows.
|
|
2384
2394
|
|
|
2385
2395
|
### POST /api/v1/jots/search
|
|
2386
2396
|
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: jaz-cli
|
|
3
|
-
version: 5.30.
|
|
3
|
+
version: 5.30.3
|
|
4
4
|
description: >-
|
|
5
5
|
Use this skill when running Clio CLI commands, building shell scripts with
|
|
6
6
|
Clio, debugging auth issues, understanding --json output, paginating results,
|
|
@@ -158,7 +158,7 @@ Rules:
|
|
|
158
158
|
| `--all` | List/search commands | Auto-paginate all pages |
|
|
159
159
|
| `--max-rows <n>` | With `--all` | Cap total rows (default 10,000) |
|
|
160
160
|
| `--finalize` | Create commands | Approve immediately (skip draft) |
|
|
161
|
-
| `--jot <text>` | Write commands (create/update/delete/pay/finalize/…) | Log the judgment behind this write in one line, inline (piggybacks a judgment-journal entry after the write succeeds; optional leading kind, e.g. `"MATCH: …"`). Without it, a successful write prints a one-line reminder to stderr — silence with `JAZ_JOTS_NUDGES=0`. |
|
|
161
|
+
| `--jot <text>` | Write commands (create/update/delete/pay/finalize/…) | Log the judgment behind this write in one line, inline (piggybacks a judgment-journal entry after the write succeeds; optional leading kind, e.g. `"MATCH: …"`). Quick LOW/MEDIUM one-liners only — for HIGH or CRITICAL calls, or when the why matters, use `clio jots create` (doctrine in its `--help`: tier anchors, kind boundaries, style). Without it, a successful write prints a one-line reminder to stderr — silence with `JAZ_JOTS_NUDGES=0`. |
|
|
162
162
|
| `--date <YYYY-MM-DD>` | Create/update commands | Transaction date |
|
|
163
163
|
| `--due <YYYY-MM-DD>` | Create/update commands | Due date |
|
|
164
164
|
| `--query <expression>` | Search commands (14 entities) | Jaz search expression (see below) |
|