PyPI - deeprefine-cli - Versions diffs - 0.1.6__tar.gz → 0.1.7__tar.gz - Mend

deeprefine-cli 0.1.6tar.gz → 0.1.7tar.gz

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.

Files changed (26) hide show

{deeprefine_cli-0.1.6/deeprefine_cli.egg-info → deeprefine_cli-0.1.7}/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deeprefine-cli
-Version: 0.1.6
+Version: 0.1.7
 Summary: CLI and Cursor skill to refine graphify knowledge graphs with DeepRefine
 Author: HKUST-KnowComp
 License-Expression: MIT
@@ -20,17 +20,15 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
-License-File: LICENSE
 Requires-Dist: openai>=1.0
-Dynamic: license-file
 <div align="center">
 # DeepRefine-Skill
-[![PyPi](https://img.shields.io/badge/PyPi-v0.1.6-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.6/)
-[![Python](https://img.shields.io/badge/Python-3.10,3.11,3.12-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.6/)
+[![PyPi](https://img.shields.io/badge/PyPi-v0.1.7-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.7/)
+[![Python](https://img.shields.io/badge/Python-3.10,3.11,3.12-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.7/)
 [![Paper](https://img.shields.io/badge/Paper-DeepRefine-b31b1b.svg)](https://arxiv.org/pdf/2605.10488)
 [![Project](https://img.shields.io/badge/Project-DeepRefine-green.svg)](https://github.com/HKUST-KnowComp/DeepRefine)
@@ -38,13 +36,13 @@ Dynamic: license-file
 Type `/deeprefine` in your AI coding assistant after you've built a **[graphify](https://github.com/safishamsi/graphify)** knowledge base — it patches `graphify-out/graph.json` from your session's query history to evolve your LLM-Wiki.
-Works in **Cursor** (install the skill once with `deeprefine cursor install`). In Cursor, `/deeprefine` tells the agent to run **`deeprefine refine`** in your project — the full Reafiner loop (judgement → abduction → action generation), not a hand-edited `graph.json`.
+Works in **Cursor** (`deeprefine cursor install`). **`/deeprefine`** runs the **same control flow as `Reafiner.refine()`** in the agent: `graphify query` / k-hop reads (no FAISS), session LLM for judgement/abduction/actions, mandatory `loop_trace_*.json`, then `deeprefine apply`. Optional **`deeprefine refine`** = full FAISS + API runtime.
 ```
 /deeprefine
 ```
-**Typical flow:** `graphify .` → `graphify query "..."` → `/deeprefine` (agent runs `deeprefine refine` for session queries).
+**Typical flow:** `graphify .` → `graphify query "..."` → `/deeprefine` (agent loop: search → judge → abduct → actions → `deeprefine apply`).
 That's it. Under `graphify-out/.deeprefine/` you get:
@@ -58,13 +56,13 @@ graphify-out/
 ```
 > **Standalone repo.** Model code (`autorefiner`, `atlas_rag`) lives in a separate [DeepRefine](https://github.com/HKUST-KnowComp/DeepRefine) checkout.
-> `pip install deeprefine-cli` ships the CLI and `SKILL.md`. `/deeprefine` and terminal `deeprefine refine` both use the same DeepRefine runtime (vLLM or API keys in your shell). Configure inference once in `atlastune`.
+> `pip install deeprefine-cli` ships the CLI and `SKILL.md`. **Agent mode** (`/deeprefine`): same loop logic, no FAISS retriever, no extra API keys. **CLI mode** (`deeprefine refine`): full `Reafiner` + embeddings (vLLM or API).
 ---
 ## News
-- [2026/6/2] deeprefine-cli v0.1.6 has been released! Customize your LLM api in CLI.
+- [2026/6/2] deeprefine-cli v0.1.7 — agent Reafiner loop + `deeprefine loop` / `apply --trace-file`.
 ---
@@ -76,7 +74,7 @@ graphify-out/
 | 2 | `pip install deeprefine-cli` |
 | 3 | `deeprefine cursor install` in your KB project |
 | 4 | (Optional) start local vLLM, or use your API provider |
-| 5 | Cursor chat: `/deeprefine` *(agent runs `deeprefine refine`)*; or terminal: `history add` + `refine` |
+| 5 | Cursor: `/deeprefine` *(agent loop)*; or terminal: `history add` + `deeprefine refine` |
 ```bash
 # 1) DeepRefine (once)
@@ -105,9 +103,9 @@ export DEEPREFINE_EMBED_MODEL=your_embed_model
 # export DEEPREFINE_API_KEY=your_shared_api_key
 # optional model overrides:
-# 5) Refine — full Reafiner loop
-# Cursor chat: type /deeprefine (agent runs deeprefine refine --query "..." per session question)
-# Terminal manual flow:
+# 5) Refine
+# Cursor: /deeprefine  (agent loop: graphify query + judgement/abduction/actions + deeprefine apply)
+# Terminal CLI mode:
 deeprefine history add --query "your question"
 deeprefine refine
 ```
@@ -255,8 +253,8 @@ deeprefine cursor install
 |:-:|--------|
 | 1 | `graphify .` or `/graphify .` → `graphify-out/graph.json` |
 | 2 | `graphify query "..."` |
-| 3 | Cursor chat: `/deeprefine` *(agent runs `deeprefine refine` / `--query` for session questions)* |
-| 4 | Terminal alternative: `deeprefine history add --query "..."` → `deeprefine refine` |
+| 3 | Cursor chat: `/deeprefine` *(agent loop; no manual history add)* |
+| 4 | Terminal: `deeprefine history add` → `deeprefine refine` *(FAISS + API/vLLM)* |
 | 5 | *(optional)* `graphify query "..."` to verify |
 ---
@@ -273,6 +271,8 @@ All commands below run from **KB project root**.
 | `deeprefine refine` | Refine all pending queries |
 | `deeprefine refine --query "..."` | Refine one query (also recorded) |
 | `deeprefine refine --rebuild-index` | Rebuild FAISS before refine |
+| `deeprefine loop init/validate/finish` | Agent loop trace (enforces Reafiner rules) |
+| `deeprefine apply --trace-file T --refinement-file F` | Apply actions (requires valid trace) |
 | `deeprefine index --rebuild` | Rebuild FAISS cache only |
 | `deeprefine cursor install \| uninstall` | Manage Cursor skill |

{deeprefine_cli-0.1.6 → deeprefine_cli-0.1.7}/README.md RENAMED Viewed

@@ -3,8 +3,8 @@
 # DeepRefine-Skill
-[![PyPi](https://img.shields.io/badge/PyPi-v0.1.6-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.6/)
-[![Python](https://img.shields.io/badge/Python-3.10,3.11,3.12-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.6/)
+[![PyPi](https://img.shields.io/badge/PyPi-v0.1.7-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.7/)
+[![Python](https://img.shields.io/badge/Python-3.10,3.11,3.12-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.7/)
 [![Paper](https://img.shields.io/badge/Paper-DeepRefine-b31b1b.svg)](https://arxiv.org/pdf/2605.10488)
 [![Project](https://img.shields.io/badge/Project-DeepRefine-green.svg)](https://github.com/HKUST-KnowComp/DeepRefine)
@@ -12,13 +12,13 @@
 Type `/deeprefine` in your AI coding assistant after you've built a **[graphify](https://github.com/safishamsi/graphify)** knowledge base — it patches `graphify-out/graph.json` from your session's query history to evolve your LLM-Wiki.
-Works in **Cursor** (install the skill once with `deeprefine cursor install`). In Cursor, `/deeprefine` tells the agent to run **`deeprefine refine`** in your project — the full Reafiner loop (judgement → abduction → action generation), not a hand-edited `graph.json`.
+Works in **Cursor** (`deeprefine cursor install`). **`/deeprefine`** runs the **same control flow as `Reafiner.refine()`** in the agent: `graphify query` / k-hop reads (no FAISS), session LLM for judgement/abduction/actions, mandatory `loop_trace_*.json`, then `deeprefine apply`. Optional **`deeprefine refine`** = full FAISS + API runtime.
 ```
 /deeprefine
 ```
-**Typical flow:** `graphify .` → `graphify query "..."` → `/deeprefine` (agent runs `deeprefine refine` for session queries).
+**Typical flow:** `graphify .` → `graphify query "..."` → `/deeprefine` (agent loop: search → judge → abduct → actions → `deeprefine apply`).
 That's it. Under `graphify-out/.deeprefine/` you get:
@@ -32,13 +32,13 @@ graphify-out/
 ```
 > **Standalone repo.** Model code (`autorefiner`, `atlas_rag`) lives in a separate [DeepRefine](https://github.com/HKUST-KnowComp/DeepRefine) checkout.
-> `pip install deeprefine-cli` ships the CLI and `SKILL.md`. `/deeprefine` and terminal `deeprefine refine` both use the same DeepRefine runtime (vLLM or API keys in your shell). Configure inference once in `atlastune`.
+> `pip install deeprefine-cli` ships the CLI and `SKILL.md`. **Agent mode** (`/deeprefine`): same loop logic, no FAISS retriever, no extra API keys. **CLI mode** (`deeprefine refine`): full `Reafiner` + embeddings (vLLM or API).
 ---
 ## News
-- [2026/6/2] deeprefine-cli v0.1.6 has been released! Customize your LLM api in CLI.
+- [2026/6/2] deeprefine-cli v0.1.7 — agent Reafiner loop + `deeprefine loop` / `apply --trace-file`.
 ---
@@ -50,7 +50,7 @@ graphify-out/
 | 2 | `pip install deeprefine-cli` |
 | 3 | `deeprefine cursor install` in your KB project |
 | 4 | (Optional) start local vLLM, or use your API provider |
-| 5 | Cursor chat: `/deeprefine` *(agent runs `deeprefine refine`)*; or terminal: `history add` + `refine` |
+| 5 | Cursor: `/deeprefine` *(agent loop)*; or terminal: `history add` + `deeprefine refine` |
 ```bash
 # 1) DeepRefine (once)
@@ -79,9 +79,9 @@ export DEEPREFINE_EMBED_MODEL=your_embed_model
 # export DEEPREFINE_API_KEY=your_shared_api_key
 # optional model overrides:
-# 5) Refine — full Reafiner loop
-# Cursor chat: type /deeprefine (agent runs deeprefine refine --query "..." per session question)
-# Terminal manual flow:
+# 5) Refine
+# Cursor: /deeprefine  (agent loop: graphify query + judgement/abduction/actions + deeprefine apply)
+# Terminal CLI mode:
 deeprefine history add --query "your question"
 deeprefine refine
 ```
@@ -229,8 +229,8 @@ deeprefine cursor install
 |:-:|--------|
 | 1 | `graphify .` or `/graphify .` → `graphify-out/graph.json` |
 | 2 | `graphify query "..."` |
-| 3 | Cursor chat: `/deeprefine` *(agent runs `deeprefine refine` / `--query` for session questions)* |
-| 4 | Terminal alternative: `deeprefine history add --query "..."` → `deeprefine refine` |
+| 3 | Cursor chat: `/deeprefine` *(agent loop; no manual history add)* |
+| 4 | Terminal: `deeprefine history add` → `deeprefine refine` *(FAISS + API/vLLM)* |
 | 5 | *(optional)* `graphify query "..."` to verify |
 ---
@@ -247,6 +247,8 @@ All commands below run from **KB project root**.
 | `deeprefine refine` | Refine all pending queries |
 | `deeprefine refine --query "..."` | Refine one query (also recorded) |
 | `deeprefine refine --rebuild-index` | Rebuild FAISS before refine |
+| `deeprefine loop init/validate/finish` | Agent loop trace (enforces Reafiner rules) |
+| `deeprefine apply --trace-file T --refinement-file F` | Apply actions (requires valid trace) |
 | `deeprefine index --rebuild` | Rebuild FAISS cache only |
 | `deeprefine cursor install \| uninstall` | Manage Cursor skill |

deeprefine_cli-0.1.7/SKILL.md ADDED Viewed

@@ -0,0 +1,311 @@
+---
+name: deeprefine
+description: >-
+  Agent-native DeepRefine Reafiner loop — same control flow as Reafiner.refine(),
+  graphify search instead of FAISS, session LLM, deeprefine apply for graph writes.
+disable-model-invocation: false
+---
+# DeepRefine — Agent Reafiner loop (strict)
+You **MUST** implement the **same control flow** as `Reafiner.refine()` in DeepRefine (`autorefiner/src/reafiner.py`).
+| Component | Agent mode | CLI `deeprefine refine` |
+|-----------|------------|-------------------------|
+| Retrieval | `graphify query` + k-hop from `graph.json` | FAISS retriever |
+| LLM | **Your session model** | External API / vLLM |
+| Graph writes | `deeprefine apply` + validated `loop_trace_*.json` | In-process |
+---
+## FORBIDDEN (hard stop)
+Do **NOT**:
+1. Run `deeprefine refine` (unless the user explicitly asks for CLI/FAISS mode).
+2. Call `deeprefine apply` without a valid `loop_trace_<query_id>.json` (CLI will reject).
+3. Skip any hop’s `<judge>Yes</judge>` / `<judge>No</judge>` judgement.
+4. Skip error abduction when `len(interaction_history) > 1`.
+5. Write `<refinement>` before abduction when refinement is required.
+6. Hand-edit `graph.json` with Python or ad-hoc JSON patches.
+7. Run `deeprefine history add` at the start of `/deeprefine` (only `loop finish` updates history).
+8. Invent a shorter pipeline (“read file → write refinement → apply”).
+If validation fails, **fix the trace and re-run the missing step** — do not bypass with `--skip-trace-check`.
+---
+## Constants (match `refine_runner.py` / Reafiner)
+```text
+MAX_HOPS = 4
+INCREMENT_HOP = 1
+BASE_TOP_K = 10
+MAX_TRIPLE_NUM_BY_STEP = [5, 10, 15, 20]   # cap triples per step
+HISTORY_HORIZON = 4                        # abduction uses last N steps
+```
+---
+## Mandatory artifact
+For each query, maintain:
+`graphify-out/.deeprefine/loop_trace_<query_id>.json`
+Create template:
+```bash
+deeprefine loop init --query "<exact question>"
+```
+Append each hop to `interaction_history` **before** starting the next hop.
+Run `deeprefine loop validate --trace-file ...` after abduction and before `apply`.
+---
+## Control flow (must match `Reafiner.refine()`)
+Pseudocode — follow **exactly**:
+```text
+interaction_history = []
+for step in 1..MAX_HOPS:
+    print "[Step: {step}]"   # show in chat
+    if step == 1:
+        # Vector-retrieval equivalent: graphify query on full question
+        RUN: graphify query "<question>"
+        triples = parse NODE/EDGE → [{subject, relation, object}, ...]
+        cap = MAX_TRIPLE_NUM_BY_STEP[0]  # 5
+        record retrieval.method = "graphify_query"
+    else:
+        # k-hop expansion from entities in previous hop (NOT a new random search)
+        entities = unique subjects/objects from previous triples
+        expand 1-hop neighbors from graphify-out/graph.json (or graphify query on entities)
+        cap = MAX_TRIPLE_NUM_BY_STEP[step-1]
+        record retrieval.method = "k_hop_expansion" or "graphify_query+k_hop_expansion"
+    triples = dedupe; len(triples) <= cap
+    # Answerable judgement — session LLM, prompts below
+    answerable, judgement_raw = LLM_judge(question, triples)
+    MUST output ONLY: <judge>Yes</judge> or <judge>No</judge>
+    append interaction_history with:
+      step, query, num_hops=(step-1)*INCREMENT_HOP, base_top_k=10,
+      retrieved_subgraph, answerable, judgement_raw,
+      retrieval: {method, evidence: "<command output excerpt>"}
+    if answerable:
+        BREAK   # stop hop loop
+# --- same branch as Reafiner.refine() line 314+ ---
+if len(interaction_history) <= 1:
+    # Early exit: first hop was answerable — NO graph refinement
+    set trace.early_exit = true
+    deeprefine loop finish --trace-file ...   # no --refinement-file
+    STOP
+else:
+    # len > 1 → ALWAYS error abduction + actions (even if last hop was Yes)
+    error_abduction = LLM_abduction(interaction_history[-HISTORY_HORIZON:])
+    MUST output: <abduction>...</abduction>
+    actions = LLM_kg_refinement(
+        last_hop.retrieved_subgraph,
+        error_abduction,
+        question,
+        source file hints from triples,
+    )
+    MUST output: <refinement>insert_edge(...)|...</refinement>
+    save refinement to graphify-out/.deeprefine/refinement_actions_<id>.txt
+    deeprefine loop validate --trace-file ... --refinement-file ...
+    deeprefine apply --trace-file ... --refinement-file ...
+    deeprefine loop finish --trace-file ... --refinement-file ...
+```
+**Critical Reafiner rule:** refinement runs when `len(interaction_history) > 1`, not only when all judgements are `No`.
+---
+## LLM prompts (verbatim — do not paraphrase)
+### Judgement (`_answerable_judgement`)
+**System:**
+```text
+As an advanced judgement assistant, your task is to judge whether the given question is answerable based on the provided KG context.
+Evaluate whether the given question is answerable based on the provided KG context. Output your judgment in the following format:
+<judge>Yes</judge> or <judge>No</judge>
+**Important:** You must think carefully about the question and the KG context before making your judgment. And output your judgment result directly in the specified format.
+```
+**User:**
+```text
+Question: {question}
+Knowledge Graph (KG) context: {triples_string}
+```
+`{triples_string}` = one triple per line: `subject | relation | object`
+### Error abduction (`_error_abduction`) — only if `len(interaction_history) > 1`
+**System:**
+```text
+As an advanced error abduction assistant, your task is to analyze the error reasons based on the given interaction history.
+Analyze the reasons of the unanswerable questions based on the given interaction history from the incompleteness, incorrectness, and redundancy perspectives. Output your analysis in the following format:
+<abduction>...</abduction>
+**Important:** You must think carefully about the interaction history before making your analysis. And output your analysis result directly in the specified format.
+```
+**User:**
+```text
+Interaction history: {interaction_history}
+```
+`{interaction_history}` format (same as Reafiner):
+```text
+Step1:
+['Query': ..., 'Subgraph_hop': ..., 'Subgraph_content': ..., 'Answerable': ...]
+Step2:
+...
+```
+### KG refinement actions (`_kg_refinement_action`) — only if `len(interaction_history) > 1`
+**System:**
+```text
+As an advanced knowledge graph refinement assistant, your task is to generate a series of actions (**within 10 actions**) to refine the given KG to make it more suitable for answering the given question.
+Based on the given KG and the analysed error reasons, refine the given KG to make it more easily for retrieval and answering the given question. You have the following three types of actions to conduct:
+- insert_edge(subject, relation, object): Insert a new edge into the KG to complete the missing information.
+- delete_edge(subject, relation, object): Delete an edge from the KG to remove the redundant information or conflicting information.
+- replace_node(old_entity, new_entity): Replace an entity in the KG to correct the errors or deal with disambiguation.
+Output a series of actions (**within 10 actions**) in the following format:
+<refinement>insert_edge("...", "...", "...")|delete_edge("...", "...", "...")|replace_node("...", "...")|...</refinement>
+**Important:** You must think carefully about the given KG and the analysed error reasons before making your refinement. DO NOT DELETE ANY IRRELEVANT TRIPLES FROM THE ORIGINAL KG. TRY TO KEEP THE ORIGINAL KG AS MUCH AS POSSIBLE. DO NOT GENERATE TOO MANY ACTIONS. And output your refinement result directly in the specified format.
+```
+**User:**
+```text
+Original Text: {original_text}
+KG: {triples_string}
+Question: {question}
+Error reasons: {error_reasons}
+```
+Use **last hop’s** `retrieved_subgraph` as `{triples_string}` (JSON list is OK in trace; string form for the prompt).
+---
+## Per-query checklist (report in chat)
+Copy and tick each item in your final message:
+```text
+[ ] Backup: graphify-out/.deeprefine/graph.json.bak
+[ ] loop_trace_<id>.json created
+[ ] Step 1: graphify query executed (evidence in trace)
+[ ] Each step: <judge>Yes|No</judge> shown in chat
+[ ] Hops stopped on Yes OR reached MAX_HOPS
+[ ] early_exit OR (abduction + refinement) per Reafiner branch
+[ ] deeprefine loop validate passed
+[ ] deeprefine apply (only if refinement path)
+[ ] deeprefine loop finish
+```
+---
+## Commands (in order)
+```bash
+# 0. KB project root; graphify-out/graph.json exists
+mkdir -p graphify-out/.deeprefine
+cp graphify-out/graph.json graphify-out/.deeprefine/graph.json.bak
+deeprefine loop init --query "<question>"
+# 1–4. For each hop: graphify/graph read → judge → append to loop_trace_*.json
+# 5a. Early exit (len(history)==1 and answerable)
+deeprefine loop validate --trace-file graphify-out/.deeprefine/loop_trace_<id>.json
+deeprefine loop finish --trace-file graphify-out/.deeprefine/loop_trace_<id>.json
+# 5b. Refinement path (len(history)>1)
+deeprefine loop validate --trace-file ... --refinement-file graphify-out/.deeprefine/refinement_actions_<id>.txt
+deeprefine apply --trace-file ... --refinement-file graphify-out/.deeprefine/refinement_actions_<id>.txt
+deeprefine loop finish --trace-file ... --refinement-file graphify-out/.deeprefine/refinement_actions_<id>.txt
+```
+---
+## `loop_trace_*.json` schema
+```json
+{
+  "schema_version": 1,
+  "mode": "agent-loop",
+  "query": "what is dulce?",
+  "query_id": "5cdc0798eb59b486",
+  "constants": { "max_hops": 4, "increment_hop": 1, "base_top_k": 10,
+    "max_triple_num_by_step": [5, 10, 15, 20], "history_horizon_size": 4 },
+  "interaction_history": [
+    {
+      "step": 1,
+      "num_hops": 0,
+      "base_top_k": 10,
+      "query": "what is dulce?",
+      "retrieval": {
+        "method": "graphify_query",
+        "evidence": "graphify query \"what is dulce?\" → …"
+      },
+      "retrieved_subgraph": [
+        {"subject": "A", "relation": "r", "object": "B"}
+      ],
+      "answerable": false,
+      "judgement_raw": "<judge>No</judge>"
+    }
+  ],
+  "error_abduction_reason": "…",
+  "error_abduction_raw": "<abduction>…</abduction>",
+  "refinement_action_file": "graphify-out/.deeprefine/refinement_actions_<id>.txt",
+  "early_exit": false
+}
+```
+---
+## Optional: CLI mode (FAISS + API)
+Only when the user **explicitly** requests `deeprefine refine` / full runtime:
+```bash
+conda activate atlastune
+export DEEPREFINE_EMBED_URL=... DEEPREFINE_LLM_URL=...
+deeprefine refine --query "..."
+```
+---
+## Paths
+- `graphify-out/graph.json`
+- `graphify-out/.deeprefine/loop_trace_*.json` (**required**)
+- `graphify-out/.deeprefine/refinement_actions_*.txt`
+- `graphify-out/.deeprefine/refinement_results_*.jsonl`
+- `graphify-out/.deeprefine/graph.json.bak`

{deeprefine_cli-0.1.6 → deeprefine_cli-0.1.7/deeprefine_cli.egg-info}/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deeprefine-cli
-Version: 0.1.6
+Version: 0.1.7
 Summary: CLI and Cursor skill to refine graphify knowledge graphs with DeepRefine
 Author: HKUST-KnowComp
 License-Expression: MIT
@@ -20,17 +20,15 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
-License-File: LICENSE
 Requires-Dist: openai>=1.0
-Dynamic: license-file
 <div align="center">
 # DeepRefine-Skill
-[![PyPi](https://img.shields.io/badge/PyPi-v0.1.6-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.6/)
-[![Python](https://img.shields.io/badge/Python-3.10,3.11,3.12-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.6/)
+[![PyPi](https://img.shields.io/badge/PyPi-v0.1.7-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.7/)
+[![Python](https://img.shields.io/badge/Python-3.10,3.11,3.12-blue.svg)](https://pypi.org/project/deeprefine-cli/0.1.7/)
 [![Paper](https://img.shields.io/badge/Paper-DeepRefine-b31b1b.svg)](https://arxiv.org/pdf/2605.10488)
 [![Project](https://img.shields.io/badge/Project-DeepRefine-green.svg)](https://github.com/HKUST-KnowComp/DeepRefine)
@@ -38,13 +36,13 @@ Dynamic: license-file
 Type `/deeprefine` in your AI coding assistant after you've built a **[graphify](https://github.com/safishamsi/graphify)** knowledge base — it patches `graphify-out/graph.json` from your session's query history to evolve your LLM-Wiki.
-Works in **Cursor** (install the skill once with `deeprefine cursor install`). In Cursor, `/deeprefine` tells the agent to run **`deeprefine refine`** in your project — the full Reafiner loop (judgement → abduction → action generation), not a hand-edited `graph.json`.
+Works in **Cursor** (`deeprefine cursor install`). **`/deeprefine`** runs the **same control flow as `Reafiner.refine()`** in the agent: `graphify query` / k-hop reads (no FAISS), session LLM for judgement/abduction/actions, mandatory `loop_trace_*.json`, then `deeprefine apply`. Optional **`deeprefine refine`** = full FAISS + API runtime.
 ```
 /deeprefine
 ```
-**Typical flow:** `graphify .` → `graphify query "..."` → `/deeprefine` (agent runs `deeprefine refine` for session queries).
+**Typical flow:** `graphify .` → `graphify query "..."` → `/deeprefine` (agent loop: search → judge → abduct → actions → `deeprefine apply`).
 That's it. Under `graphify-out/.deeprefine/` you get:
@@ -58,13 +56,13 @@ graphify-out/
 ```
 > **Standalone repo.** Model code (`autorefiner`, `atlas_rag`) lives in a separate [DeepRefine](https://github.com/HKUST-KnowComp/DeepRefine) checkout.
-> `pip install deeprefine-cli` ships the CLI and `SKILL.md`. `/deeprefine` and terminal `deeprefine refine` both use the same DeepRefine runtime (vLLM or API keys in your shell). Configure inference once in `atlastune`.
+> `pip install deeprefine-cli` ships the CLI and `SKILL.md`. **Agent mode** (`/deeprefine`): same loop logic, no FAISS retriever, no extra API keys. **CLI mode** (`deeprefine refine`): full `Reafiner` + embeddings (vLLM or API).
 ---
 ## News
-- [2026/6/2] deeprefine-cli v0.1.6 has been released! Customize your LLM api in CLI.
+- [2026/6/2] deeprefine-cli v0.1.7 — agent Reafiner loop + `deeprefine loop` / `apply --trace-file`.
 ---
@@ -76,7 +74,7 @@ graphify-out/
 | 2 | `pip install deeprefine-cli` |
 | 3 | `deeprefine cursor install` in your KB project |
 | 4 | (Optional) start local vLLM, or use your API provider |
-| 5 | Cursor chat: `/deeprefine` *(agent runs `deeprefine refine`)*; or terminal: `history add` + `refine` |
+| 5 | Cursor: `/deeprefine` *(agent loop)*; or terminal: `history add` + `deeprefine refine` |
 ```bash
 # 1) DeepRefine (once)
@@ -105,9 +103,9 @@ export DEEPREFINE_EMBED_MODEL=your_embed_model
 # export DEEPREFINE_API_KEY=your_shared_api_key
 # optional model overrides:
-# 5) Refine — full Reafiner loop
-# Cursor chat: type /deeprefine (agent runs deeprefine refine --query "..." per session question)
-# Terminal manual flow:
+# 5) Refine
+# Cursor: /deeprefine  (agent loop: graphify query + judgement/abduction/actions + deeprefine apply)
+# Terminal CLI mode:
 deeprefine history add --query "your question"
 deeprefine refine
 ```
@@ -255,8 +253,8 @@ deeprefine cursor install
 |:-:|--------|
 | 1 | `graphify .` or `/graphify .` → `graphify-out/graph.json` |
 | 2 | `graphify query "..."` |
-| 3 | Cursor chat: `/deeprefine` *(agent runs `deeprefine refine` / `--query` for session questions)* |
-| 4 | Terminal alternative: `deeprefine history add --query "..."` → `deeprefine refine` |
+| 3 | Cursor chat: `/deeprefine` *(agent loop; no manual history add)* |
+| 4 | Terminal: `deeprefine history add` → `deeprefine refine` *(FAISS + API/vLLM)* |
 | 5 | *(optional)* `graphify query "..."` to verify |
 ---
@@ -273,6 +271,8 @@ All commands below run from **KB project root**.
 | `deeprefine refine` | Refine all pending queries |
 | `deeprefine refine --query "..."` | Refine one query (also recorded) |
 | `deeprefine refine --rebuild-index` | Rebuild FAISS before refine |
+| `deeprefine loop init/validate/finish` | Agent loop trace (enforces Reafiner rules) |
+| `deeprefine apply --trace-file T --refinement-file F` | Apply actions (requires valid trace) |
 | `deeprefine index --rebuild` | Rebuild FAISS cache only |
 | `deeprefine cursor install \| uninstall` | Manage Cursor skill |

{deeprefine_cli-0.1.6 → deeprefine_cli-0.1.7}/deeprefine_cli.egg-info/SOURCES.txt RENAMED Viewed

@@ -12,6 +12,9 @@ deeprefine_cli.egg-info/top_level.txt
 deeprefine_skill/SKILL.md
 deeprefine_skill/__init__.py
 deeprefine_skill/adapter_graphify.py
+deeprefine_skill/agent_graph.py
+deeprefine_skill/agent_loop.py
+deeprefine_skill/agent_prompts.py
 deeprefine_skill/cli.py
 deeprefine_skill/history.py
 deeprefine_skill/installers.py

deeprefine-cli 0.1.6__tar.gz → 0.1.7__tar.gz

deeprefine-cli 0.1.6tar.gz → 0.1.7tar.gz